INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction Is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough. substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g.. maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higfwr quality 6" x 9* black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

ProQuest Information and Leaming 300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 800-521-0600 UMÏ

MEMBRANE TOPOLOGY AND STRUCTURE -FUNCTION RELATIONSHIP OF

THE APICAL SODIUM-DEPENDENT BILE ACID TRANSPORTER. ASBT

DISSERTATION

Presented In Partial Fulfillment of the Requirements for the

Degree Doctor of Philosophy In the Graduate School of

The Ohio State University

By

Yongheng Zhang, M.S.

****

The Ohio State University

2001

Dissertation Committee: Approved by Dr. Peter Swaan, advisor

Dr. William Hayton

Dr. Daren Knoell Advisor

Dr. James Dalton College of Pharmacy UMI Number: 3031293

UMI*

UMI Microform 3031293 Copyright 2002 by Beil & Howell Information and Leaming Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

Bell & Howell Information and Leaming Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 © Yongheng Zhang 2001 All Rights Reserved ABSTRACT

The apical sodium-dependent bile acid transporter (ASBT) is a member of

the solute carrier transporter superfamily and an essential component of

enterohepatic circulation. ASBT has been investigated as a vector to improve

oral drug bioavailability and appreciated as a novel molecular target for

cholesterol-lowering therapy. This dissertation is aimed to (I) to define the

membrane topology of ASBT, and (II) to identify critical amino acid residues

involved in sodium and bile acid interaction.

Chapter 1 presents an overview of membrane transporters and current transporter research. This part includes a discussion on pharmaceutical

relevance of transporters, transporter classification and examples to study transporter function. In the second part, a detailed literature review on ASBT is presented.

In Chapter 2, experimental strategies to address Aim I are delineated. In short, /V-glycosylation sites are engineered into the strategic sites on the non­ glycosylated form of ASBT (N10D mutant) to differentiate between the two proposed topology models (7 versus 9 transmembrane domains). Results indicate that /V-glycosylation sites are successfully introduced into the strongly supporting a 7TM model for ASBT.

In Chapter 3, approaches to address Aim II are described. Based on the

defined membrane topology (7TM), alanine scanning mutagenesis is subsequently applied to 12 highly conserved charged residues in the extracellular domains of ASBT to generate a mutant library; expression and function ([^H]-taurocholic acid (TCA) uptake kinetics and sodium dependency) of wild-type and mutants are analyzed by western blot and uptake studies in the transiently transfected COS -1 cells; uptake inhibition studies with various naturally occurring bile acid analogues are performed to further structurally probe the transporter/substrate interaction. Results show that [^H] TCA uptake kinetics is significantly altered in a single functional mutant, D124A, which additionally has decreased affinity to other bile acid analogues containing a 7a hydroxyl group. Sodium activation kinetics indicates a decreased apparent sodium affinity for mutant R254A. In conclusion, kinetic studies of ASBT mutants reveal a probable region for bile acid recognition, Asp124 (on extracellular loop 1), and a candidate region for sodium interaction/translocation, Arg254 (on extracellular loop 3).

In Chapter 4, future directions to extend this dissertation work are presented.

Ill DEDICATION

To my family

IV ACKNOWLEDGMENTS

I would like to extend my gratitude to my advisor, Dr. Peter Swaan, for

scientific guidance, encouragement and financial support throughout my graduate study, and for his patience in polishing my presentation and scientific writing skills.

I would like to thank Dr. Daren Knoell for inspiring discussions and for providing resources that made this thesis possible. I would also like to thank my committee members. Dr. William Hayton and Dr. James Dalton for their time and valuable suggestions.

I am indebted to my labmates: Se-ne Huang, Mitch Phelps, Cheng Chang,

Antara Banerjee, Amy Foraker and Hulya Ulukan. Many thanks to Kathy Brooks and Karen Lawler. It is their friendship and timely help that made my graduate life both enjoyable and productive.

I am grateful to my parents and parents in-iaw for their unconditional support and understanding; to my wife and son for their love and being part of my life. VITA

September 25,1972 ...... Bom - Ulumuqi, China

1993 ...... 8 .8 . Pharmacy

China Pharmaceutical University

1996...... M.S. Pharmceutical Analysis

China Pharmaceutical University

1996 - present...... Graduate Teaching and Research

Associate, The Ohio State University

PUBLICATIONS

1. E. Y. Zhang, M. Phelps and P. Swaan. Essential residues in the apicai

sodium-dependent bile acid transporter (ASBT) for sodium and substrate

interaction. Submitted.

2. E. Y. Zhang, M. Phelps, F. Helsper and P. Swaarv. Membrane topology and

putative structure of the apical sodium-dependent bile acid transporter

(ASBT). Submitted.

3. E.Y. Zhang, M. Phelps, C. Chang, S. Ekins and P. Swaan. Modeling of active

transport systems. Advanced Drug Delivery Reviews. In press.

VI “4r Er-Yr“ZhangrQrKnlpprSrEkins-and-PrSwaanrStrticturahbiotogy*and-functiofr

of solute transporters: Implications for identifying and designing substrates.

Drug Metabolism Reviews. In press.

5. E. Y. Zhang, F. Helsper and P. Swaan. Use of the apical sodium-dependent

bile acid transporter (ASBT) for drug delivery. Drug Targeting to the

Gastrointestinal Tract. Harwood Academic. In press.

6. E. V. Zhang and P. Swaan. Determination of membrane protein glycation in

diabetic tissue. AAPS PharmSci 1999; 1 (4) article 20

fhttD://www.Dharmsci.ora/scientificioumals/Dharmsci/iournal/99 20.htmh.

7. E. V. Zhang and P. Swaan. Changes in intestinal barrier function and active

transport during uncontrolled diabetes. FASEBJ. 13: A i d 2 1999.

8 . E. V. Zhang and P. Swaan. A colorimetric assay for determining the glycation

level of the total membrane proteins. PharmSci 1 (IS ) 232 1998.

FIELDS OF STUDY

Major Field: Pharmacy

Studies in Pharmaceutics and Drug Transporters

VII TABLE OF CONTENTS

ABSTRACT...... Il DEDICATION...... IV ACKNOWLEDGMENTS...... V VITA ...... VI LIST OF FIGURES...... XI LIST OF TABLES...... XIII

CHAPTER 1 ...... 1 INTRODUCTION...... 1 1.1 Overview on membrane transporters...... 1 1.1.1 Pharmaceutical relevance of membrane transporters...... 3 1.1.2 Classification of transporters...... 3 1.1.3 Approaches to obtain structural information of transporters...... 5 1.2 Apical sodium-dependent bile acid transporter...... 14 1.2.1 Enterohepatic circulation (EMC) and the bile acid transporters ...... 15 1.2.2 Sodium-dependent bile acid cotransporters (SBATs)...... 18 1.2.3 Expression, regulation and transport properties of the ASBT...... 19 1.2.4 Structure - activity relationship (SAR) for A SB T ...... 22 1.2.5 Use of bile acid transport system for drug delivery ...... 25 1.2.6 Summary ...... 28 1.3 Significance of this dissertation work ...... 29

CHAPTER 2 ...... 57 MEMBRANE TOPOLOGY OF ASBT...... 57

VIII I “ -—2r1~lfrtfOClUCtion ...... Trnrrr» : 7. .7.7::— : ;. — n.. r.v; rr.v.T..%...... i v . .... r.-v;. 2.2 Experimental procedures ...... 59 2.2.1 Materials.;...... 59 2.2.2 Cell culture ...... 60 2.2.3 Antibody preparation ...... 60 2.2.4 Site-directed mutagenesis...... 60 2.2.5 Transient transfection of ASBT in COS-1 cell ...... 61 2.2.6 Uptake assay, western blot and cell surface biotinylation ...... 63 2.2.7 Data Analysis ...... 64 2.3 Results...... 64 2.3.1 Transient transfection...... 64 2.3.2 AAlinked glycosylation in ASBT ...... 65 2.3.3 Membrane topology of ASBT ...... 68 2.4 Discussion ...... 69

CHAPTER 3 ...... 88 RESIDUES ESSENTAIL FOR SODIUM AND BILE ACID INTERACTION IN ASBT...... '...... 88 3.1. Introduction ...... 88 3.2 Experimental procedures ...... 90 3.2.1 Materials...... 90 3.2.2 Site-directed mutagenesis...... 90 3.2.3 Cell culture andtransfection ...... 91 3.2.4 Bile acid uptake assay ...... 92 3.3 Results...... 93 3.3.1 Immunoblots...... 93 3.3.2 Uptake kinetics of mutants...... 95 3.3.3 Sodium activation kinetics ...... 97 3.4 Discussion ...... 98

IX FUTURE DIRECTIONS...... 116 4.1 Helix packing of transmembrane segments of ASBT...... 116 4.2 Further verification of biological relevance of pharmacophore models... 117 4.3 Construction of a comprehensive structural and predictive model of ASBT 118 4.4 Crystallization of ASBT or structurally related transporters ...... 120 4.5 Membrane topology and structure-functlon relationship of NTCP ...... 121 4.6 Potential protein Interaction between ASBT and ILBP...... 122 4.7 Final remarks...... 124

BIBLIOGRAPHY...... 127

APPENDIX...... 148 LIST OF FIGURES

Figure 1.1 The currently recognized primary types of human transporters ...... 32

Figure 1.2 Rationale for AAglycosylation scanning mutagenesis ...... 44

Figure 1.3 Enterohepatic circulation ...... 45

Figure 1.4 Structure and nomenclature of bile acids ...... 47

Figure 1.5 Bile acid transport system in the enterocytes ...... 49

Figure 1.6 Bile acid transport system in the hepatocytes ...... 50

Figure 1.7 Sequence alignment for human ASBT and NTCP...... 51

Figure 1.8 Hydropathy plots for human ASBT and N TC P ...... 52

Figure 1.9 Multiple sequence alignment of ASBTs...... 54

Figure 1.10 Model of transcriptional control of bile acid and cholesterol

homeostasis by nuclear receptors ...... 55

Figure 2.1 The 7TM versus 9TM membrane topology model for ASBT...... 74

Figure 2:2 The strategic locations of the inserted newrAf^giycosyiaiotn sites ..... 76

Figure 2.3 Sequences of the mutagenesis primers...... 78

Figure 2.4 ASBT has only one AAglycosylation site occurring at Asn^° ...... 80

Figure 2.5 Uptake activity of ASBTs from which the endogenous /V-glycosylation

consensus sites have been removed...... 82

Figure 2.6 Analysis of aglyco-ASBT with the engineered AAglycosylation sites 84

XI Figure 2.7 Uptake activity of aglyco-ASBTs (N10D) in which the /V-glycosylation

consensus sites have been inserted...... 86

Figure 3.1 Putative seven-TMs topology model for ASBT...... 105

Figure 3.2 Immunodetection of ASBT wild-type and Ala-substituted mutants. 106

Figure 3.3 Sodium-dependent uptake of [^H] TCA by wf and Ala-substituted

mutants...... 108

Figure 3.4 Structures of bile acid analogues...... 109

Figure 3.5 Inhibition of TCDCA on ^H-TCA uptake of wild-type andmutants .110

Figure 3.6 Sodium dependency of [^H]-TCA uptake by wt and mutants ...... 112

Figure 3.7 Sodium activation kinetics of wt and mutants ...... 113

Figure 3.8 Putative interactions between ASBT and bile acid ...... 115

Figure 4.1 Pharmacophore model for ASBT(A) and stucture for natural ligand (B)

...... 125

Figure 4.2 Putative 7TM model of Human ASBT and NTCP...... 126

XII LIST OF TABLES

Table 1.1 Top 100 PDA approved drugs...... 30

Table 1.2 Human ATP-binding cassette (ABC) transporter superfamily...... 31

Table 1.3 Overview of the Solute Carrier (SLC) protein genetic superfamily 33

Table 1.4 Techniques for studying polytopic membrane protein structure 43

Table 1.5 Bile acid transporters...... 48

Table 1.6 Sodium-dependent bile acid transporters...... 53

Table 3.1 Sequences of mutagenesis primers...... 104

Table 3.2 TCA uptake kinetic constants of wild-type and mutants ...... 107

Table 3.3 Summary of IC 50 values of each bile acid analogue to wt and mutants

...... I l l

XIII m .:

CHAPTER 1

INTRODUCTION

As a general Introduction to this dissertation, this chapter aims (1) to give

an overview on current transporter research, and (2) to provide the essential

biochemical and physiological background for the apical sodium-dependent bile

acid transporter (ASBT), which can be regarded as a general paradigm of solute

carrier proteins and Is the focus of this dissertation work.

1.1 Overview on membrane transporters

Membrane transporters In general are a large group of membrane proteins

that have one (bItopIc) or more (polytopic) hydrophobic transmembrane

segments. These transporters are Involved In almost all facets of biological

processes In the cell [1].

"(1) They allow entry of all essential nutrients Into the cytoplasmic

compartment and subsequently Into organelles, allowing metabolism of

exogenous sources of carbon, nitrogen, sulfur, and phosphorus; — - — (2)^hey^pfovide-am@ams-foi"regu*ation-of-metabolites'Goncentra*ion» by

catalyzing the excretion of end products of metabolic pathways from organelles

and cells;

(3) They mediate the active extrusion of drugs and other toxic substances

from either the cytoplasm or the plasma membrane;

(4) They mediate uptake and efflux of ion species that must be maintained

at concentrations that differ dramatically from those in the external milieu;

(5) Transporters participate in the secretion of protein, complex

carbohydrates, and lipids into and beyond the cytoplasmic membrane;

(6) Transport systems allow the transfer of nucleic acids across cell

membranes, allowing genetic exchange between organisms and thereby

promoting species diversification;

(7) Transporters facilitate the uptake and release of pheromones,

hormones, neurotransmitters, and a variety of other signaling molecules that

allow a cell to participate in the biological experience of multicellularity;

(8) Transporters allow living organisms to conduct biological warfare,

secreting, for example, antibiotics, antiviral agents, antifungal agents, and toxins

in humans and other animals that may confer upon the organisms producing

such an agent a selective advantage for sun/ival purposes. Many of these toxins

are themselves channel-forming proteins or peptides that serve a cell-disruptive

transport function.” -H-rtrl-Pharmaeeutieaf^felevanee-ofmembfane-transportefs------

From a pharmaceutical point of view, membrane transporters are

important for at least two reasons. One is that they can actively transport many

drug molecules across biological membranes, and play very important role in

drug entrance and exit from the body. Secondly, they serve as the

pharmacological targets for many drugs. In fact, the molecular targets of more

than 50% of FDA approved drugs are membrane proteins such as channels,

transporters and receptors (Table 1.1). It has been well recognized that the

absorption, distribution, metabolism, excretion and toxicity (ADME/T), and

pharmacokinetic / pharmacodynamic (PK/PD) profiles of many drug molecules

can be influenced by the expression level, function and pharmacogenomics of

drug transporters [2, 3].

1.1.2 Classification of transporters

Figure 1.1 shows the currently recognized primary types of human

transporters. These proteins are initially divided into channels and carriers.

Compared with carriers, channels transport their substrates such as ions and

small molecules in a diffusion limiting process and allow for an extremely fast

rate of transport.

Depending on the energy sources and energy coupling mechanism,

carriers can be further subdivided into three groups. Uniporters, also called

facilitated diffusion carriers, transport single substrate in a passive process driven

by the substrate's own electrochemical gradient. Primary and secondary active —tran9portersron-the‘Other-handrearhtfansport-stibstrat»-acrosrthe^rnembran&

against an electrochemical gradient. Primary active transport is driven by the

energy derived from chemical reaction such as adenosine triphosphate (ATP)

hydrolysis, light absorption or electron flow. Secondary active transport is driven

by the electrochemical gradient of ions, such as sodium and proton, created by

primary active transport systems.

This way of classifying transporters has been proven useful due to its

clarity and simplicity. However, with more and more transporters been cloned

and functionally characterized, it becomes clear that, when classifying

transporters, we need take account of many essential features of the

transporters, such as transport function, substrate specificity, sequence similarity

and evolutionary origins. In fact, there are several excellent reviews addressing

the complexity of classifying transporters [1,4].

While the humari gene nomenclature committee (HUGO,

http://www.gene.ucl.ac.uk/nomenciature/) is taking the effort to establish the

detailed guidelines to classify transporter, several distinct transporter

superfamilies have been recognized and designated. One example is the ATP-

binding cassette (ABC) transporter superfamily (Table 1.2). These transporters

bind ATP and use the energy to drive the transport of various molecules across

the plasma membrane as well as the intracellular membrane. Currently the ABC

transporter family in human has seven subfamilies with 48 members [5]. For

example, the well-known muti-drug resistance proteins, such as MDR, i.e. P-

glycoprotein (P-gp), and muti-drug resistant associated protein MRP, are inctudechin-thirtransportersaperfarTTityrAnoth9rexarnpl9 ts'thff SotutrCarrter ^

(SLC) transporter family (Table 1.3), which currently contains 37 subfamilies and

208 transporter members.These transporters are essential for the cellular uptake

and homeostasis of many essential nutrients.

1.1.3 Approaches to obtain structural Information of transporters

The knowledge of structural Information of proteins Is essential to

understanding their function. Unlike soluble proteins, hydrophobic membrane

proteins are notoriously resistant to purification and crystallization. In fact, less

than 20 high-resolutlon structures of membrane proteins have been resolved so

far, whereas more than ten thousand three-dimensional structures of soluble

globular proteins are recorded at high resolution (www.pdb.org). Because of this,

blochemlcal/blophyslcal and computational approaches are needed to obtain

structural information for membrane transporters.

Although Integral membrane proteins display a variety of functions, they

share similar architecture; they cross the membrane In a zigzag fashion, embed their hydrophobic segments In the llpid bllayer and expose their hydrophilic loops, which connect the transmembrane segments. In the outside and Inside of the membrane. Therefore, knowledge of membrane topology (I.e. the number of transmembrane segments and their orientation In the membrane) Is a fundamental aspect of the structural elucidation of membrane proteins. For a specific membrane transporter. Its membrane topology can be first predicted by various computer software packages, which are all based on similar algorithms

[6]. However no existing method can definitively determine membrane topology residues that are suitable for the formation of a membrane spanning segment.

Examples of these software programs Include TopPred II [7]

(http://www.blokeml.su.se), the PHD topography neural network system

developed by Rost and co-workers [8]

(http://www.emblheldeelberg.de/predlcproteln), or the hidden Markov model

recently described by Tusnàdy and Simon [9] (http://www.enzlm.hu/hmmtop).

The predicted topology then needs be verified and Improved by a variety of

biochemical and biophysical techniques.

1.1.3.1 Biochemical and biophysical approaches

The rationale of these experimental approaches Is based on the aforementioned architectural principles of membrane proteins. Because the llpid bllayer represents an Impermeable barrier to hydrophilic compounds, the extracellular and Intracellular loops and the hydrophobic transmembrane domains of a membrane protein are differently accessible to various sided reagents. By using site directed mutagenesis, some easily Identified target sites, such as Cys residues [10-12], /V-glycosylation sites [13-17], antibody epitopes

[18-20], and proteolytic sites [21,22], can be Inserted Into specific strategic positions in the protein. The accessibility of these sites to the sided reagents Is then determined. For example, specific amino acids are mutated to Cys and their accessibility Is tested with thiol-reacting chemicals. Residues accessible from the extracellular side can be modified by both membrane-impermeable and -

6 -penmable^reagentSrwhereas-Feslckies-en-tlie-eytoplasmic-sjde-are^only^^ modified-

by the latter; residues that are completely buried in the membrane will not react

at all. By probing different regions of the protein by these target sites, protein

membrane topology can be determined. Besides this insertion approach, the

topological information of a protein can be also obtained through gene fusions. A

reporter molecule, e.g. an enzyme tag (alkaline phosphatase, p-galactosidase or

p-lactamase [23-27]), is fused to a hydrophilic domain of a membrane protein to

generate a fusion protein. The property of the fused reporter is then determined,

and results indicate the cellular location of the fusion site. These techniques have

all been used individually or in combination. Each of these techniques has

advantages and disadvantages, and only rarely does one method enable

conclusive topographic analysis of a polytopic integral membrane protein.

Since protein function depends on its structure, elucidation of the

structure-function relationship of a membrane protein is one of the key issues of

protein research. The functionally important residues in the protein can be

revealed by site-directed mutagenesis. For example, specific amino acids are

mutated to alanine to generate a mutant library and protein function is analyzed.

The mutants that retain the full wide type activity suggest that these mutation

sites are not essential for protein function, whereas the loss-of-function mutations

indicate that these mutation sites are likely functionally important.

Considerable progress has been made in the past 10 years by the group

of Kaback at UCLA, who has taken numerous biophysical approaches towards

studying the structure of the bacterial transporter lactose permease [28-37].

7 -Table-1^4^lis*s^*he^*e€hniques^(haMhi9-an&@*heFgr@wp9^have^wse€H»90lwthe^

structure and topology of membrane transporters In the absence of a crystal

structure. Several of these techniques are described In detail below.

1.1.3.1.1 Cys-scanning mutagenesis / substituted cysteine accessibility method

Studying the lactose permease of Escherichia coll, a polytopic membrane transport protein that catalyzes beta-galactoslde/H^ symport, Frilllngos and colleagues [38,39] used Cys-scannIng mutagenesis In order to determine which residue play an obligatory role In the transport mechanism. In their study, all of the endogenous Cys residues were mutated to Ser or Ala to generate a Cys-less mutant, and then a library of mutants with a sIngle-Cys residue at each position of the protein was created for structure/function studies. In general, these types of studies will define amino acid side chains that play an essential role In the transport mechanism and positions where the reactivity of the Cys replacement is altered upon ligand binding. Furthermore, helix packing, helix tilt, and llgand-

Induced conformational changes can be determined by using the library of mutants In conjunction with a battery of site-directed techniques. Another name for this techniques Is the substituted cysteine accessibility method (SCAM), which has been successfully used to address both protein dynamics and topology. Including P-glycoproteIn [40], 5-hydroxytryptamlne receptors [41], human A(1) adenosine receptor [42] and opioid receptor [43]. However, the application of this approach Is only applicable to membrane proteins whose Cys- less mutant Is biologically functional.

8 1.1.3.1.2 /V-giycosylation and epitope scanning mutagenesis

Since /V-glyccsylation occurs only on the luminal side of the endoplasmic

reticulum, the /V-glycosylation side chain provides a topological marker for the

extracellular side of the membrane protein (Figure 1.2). This method has been

successfully used to determine which domains of the receptors/channels are located extracellularly. In general, a glycosylation-free (aglyco) mutant is first generated by the disruption of the endogenous AAglycosylation site(s) and subsequently, AFglycosylation consensus sequences (NXS/T) are engineered into strategic regions of the aglyco-mutant. Based on the positioning of the glycosylation groups within the extracellular membrane, a molecular weight shift indicates successful glycosylation and reveals definitive information on protein topology. A drawback of this technique is the dependence of glycosylation on efficiency and accessibility of the consensus sequence to the glycosylation machinery; thus, non-glycosylated mutants do not provide conclusive information about the topology and these data should be interpreted carefully. Regardless, the topology of various proteins has been successfully solved using this technique, including the sodium dependent glucose transporter (SGLT1) [44], the

Y-aminobutyric acid transporter, GAT-1 [45], and P-giycoprotein [12].

Alternatively, small peptide epitopes can be inserted into the extramembranous parts of a transporter, which are recognized by a well- characterized monoclonal antibody. Covitz and colleagues [19] used this technique to determine the topology of the human peptide transporter, hPepTI.

9 -Arrepitope tagrEYMPMErwag mserted-into different extramembranooy tecattons-

of hPepTI by site-directed mutagenesis. The membrane topology was solved by

labeling reconstituted, functionally active, EYMPME-tagged hPEPTI mutants

with an anti-EYMPME- monoclonal antibody in non-permeabilized and

permeabilized cells.

1.1.3.1.3 Site-directed spin-labeling and excimer fluorescence

Membrane transporters generally undergo large-scale conformational

changes to fulfill their biological functions. Therefore, in addition to the

techniques that address the static' structures of membrane proteins, techniques

that allow the study of protein dynamics are also needed. Site-directed spin-

labeling (SDSL) and site-directed excimer fluorescence (SDEF) are useful

techniques in this respect. SDSL can be used to monitor (often in real time)

conformational changes and is extremely powerful in combination with structurai

information [46,47]. The strategy involves introducing nitroxide radicals at specific positions in a protein via site-specifically-placed Cys residues. Electron paramagnetic resonance spectroscopy (EPR) of the spin-labeled proteins yields specific information about protein dynamics based on the mobility of the attached nitroxide, its accessibility to collisions with lipid- or water-soluble quenchers and distance measurements between pairs of nitroxides through dipole-dipole interaction. The application of this technique was proven very useful in the determination of the local structure and dynamics of a variety of transporters [48-

50].

10 -Stte»clirected-excifnet.fUiore8C8nce-(SDEg)-»^partiailarty^usefut-to-8tu

proximity reiationships between membrane helices. The experiments are based

upon site-directed pyrene labeling of combinations of paired Cys replacements in

a mutant devoid of Cys residues. Since pyrene exhibits excimer fluorescence if

two molecules are within about 3.5 A, the proximity between paired labeled

residues can be determined. Moreover, interspin distances in the range of 8-25 A

between two spin-labeled Cys residues can be measured in the frozen state.

Using this technique, Kaback and co-workers showed that ligands of the lac

permease cause a dramatic increase in reactivity, which is consistent with the

notion that the mutated amino acid positions are transferred into a more

hydrophobic environment [51-53].

1.1.3.1.4 Site-directed chemical cleavage

The insertion of short reporter sequences, e.g. factor Xa protease cleavage sites, into hydrophilic loops is proven a useful alternative to N- glycosylation scanning mutagenesis. However, this approach requires isolation of homogeneous preparations of intact membranes and many tedious control experiments. Furthermore, experimental difficulties associated with protease accessibility are well documented [54]. In general, frame factor Xa protease sites are inserted into a target sequence at positions within the NHg- and COOH- terminal domains, and into hydrophilic loops. The factor Xa protease recognizes the tetrapeptide motif lEGR and specifically cleaves the protein sequence in the

COOH-terminai of the arginine residue [55]. Generally, the recognition motif is

11 .tandomLy.(lEGBlEGB>.insertecLtaincrease.the.pcobabUit\tof cleavage. Aftet

digestion of purified protein vesicles with the factor Xa enzyme, fragments are

isolated on SDS-PAGE and can be analyzed to further determine membrane

topology.

1.1.3.1.5 Membrane Insertion Scanning

Sachs and colleagues at UCLA pioneered a technique aptly named

“membrane insertion scanning” to determine whether a sequence of amino acids

is capable of spanning a membrane [56]. They tested the ability of individual

protein segments to function as signal anchors for membrane insertion by

placement between a cytoplasmic anchor encompassing the first 101 amino

acids of the rabbit H*, K* ATPase (HK MO) -subunit and a glycosylation flag

sequence consisting of five AAlinked glycosylation sites located in the C-terminal

177 amino acids of the rabbit HK MO-subunit. Stop transfer properties of

hydrophobic sequences can be examined in a similar manner using the first 139

N-terminal amino acids of the rabbit H* K*-ATPase -subunit (HK M l) containing

the first membrane sequence of the H \ K^-ATPase as a signal anchor upstream

of individual predicted transport transmembrane sequences linked to the N-

glycosylation flag. The membrane insertion scanning technique has been used to

determine the membrane topology of various transporters, channels and

receptors [57,58]. A common criticism of membrane insertion studies is the placement of a hydrophobic amino acid sequence out of its physiological or environmental context, thus forcing' an abbreviated sequence through the

12 -membFanerTFansmernbFane-FegkNFis-HFi-peiytopte'mefnbFane-pfoteins-may^^require-

flanking topogenic Information to become integrated into the lipid bilayer and

thus, topology determination by this technique alone may not be definitive.

1.1.3.2 In silico approaches

With the dramatic advances in computer technology and modeling

algorithms, molecular modeling is becoming a feasible alternative to predict the

structures of membrane proteins. One approach is homology modeling. It is well

recognized that any two proteins that show sequence homology (i.e., share

sufficient primary structural similarity to have evolved from a common ancestor)

will prove to exhibit strikingly similar three dimensional (30) structures [59].

Furthermore, the degree of tertiary structural similarity correlates well with the

degree of primary structural similarity. Phylogenetic analyses allow application of

modeling techniques to a large number of related proteins and additionally allow

reliable extrapolation from one protein family member of known structure to

others of unknown structure. Thus, once 3D structural data are available for any

one family member, these data can be applied to all other members within limits

dictated by their degrees of sequence similarity. The 3 0 structures of some

receptors/transporters have been predicted by this approach [60-64].

Another approach is to use the quantitative structure activity relationship

(OSAR) to construct a pharmacophore model for the receptor/transporter. Based

on the structural information and activity data of a training set of substrates for a

given receptor/transporter, the chemical features essential for the activity/affinity

13 ^ - f '

-ef-8ub6tFate‘eaFf-be-deFived-to-biHld-a-phaFmaeoplioFe-mGdel^complimentaF^t& the active site of the protein. Since at least 20-30 chemically diverse compounds need to be individually characterized (i.e., known structure and biological endpoint) to allow for a QSAR paradigm to be successfully applied, this approach is especially applicable to some important pharmaceutically relevant transporters/receptors, where reliable chemical and biological information of a relatively large set of compounds to the transporter/receptor has been gathered in the drug discovery process and hence available for assessment and prediction. Examples include the multiple drug efflux transporter P-glycoprotein and MRP [65,66], intestinal peptide transporter PepT1[67], intestinal bile acid transporter [68,69], serotonin (5-HT) receptor [70], and dopamine transporter

[71, 72].

1.2 Apical sodium-dependent bile acid transporter

As a paradigm for SLC transporters, we focus this dissertation work on the apical sodium-dependent bile acid transporter (ASBT), which plays a key role in the enterohepatic recycling of bile salts and cholesterol homeostasis. ASBT is a transporter with significant pharmaceutical relevance; It has been previously targeted for drug delivery purposes, either by mimicking endogenous substrate or prodrug approach [73,74]; besides its use in optimizing oral drug delivery, ASBT has been appreciated recently as a pharmacological target for cholesterol- lowering therapy [75, 76]. This dissertation section is presented to give a detailed review on the biochemistry and physiology of ASBT, the structure-activity

14 -relation8hip-ef-ASB-Trand-reeent-advanee8-in-u8ing-ASB1^fo^€lftig-€leli¥eFy>

purposes.

1.2.1 Enterohepatic circulation (EHC) and the bile acid transporters

Enterohepatic circulation (EHC) is highly efficient in the conservation of

bile acids (Figure 1.3). The major bile acids, cholic acid and chenodeoxycholic

acid, are biosynthesized from cholesterol, catalyzed by the rate-limiting

microsomal cholesterol 7 a-hydroxylase (CYP7A1), and conjugated with glycine

or taurine (Figure 1.4) in the liver. Conjugated bile acids are then secreted from

the liver and stored in the gallbladder during fasting. The ingestion of a meal

triggers the contraction of the gallbladder. As a result, bile enters the duodenum

and facilitates the digestion and absorption of fat and fat-soluble vitamins. Bile

acids are passively reabsorbed in the proximal part of small intestine, actively

assimilated in distal ileum, and transited to the liver through portal circulation.

After being taken up by the liver, bile acids are resecreted into bile and a new cycle begins. A 70-kg human secretes SOg/day of bile acids. Of this 30g, 0.3-0.5g is lost in the feces and this loss is recovered by the biosynthesis from cholesterol in the liver. Therefore, 98% of bile acids is reabsorbed and circulates in the EHC.

However, the total bile acid pool is only 3g. This means that the pool must circulate 10 times/day to match the secretion rate of 30g/day. Therefore, only a small synthetic rate 0.3-0.5g/day is needed to maintain a 3g-pool.

Bile acids are absorbed from the intestine through a combination of passive absorption in the proximal small intestine and active transport in the

15 -di8tat-iieumr-Th»-ileakaGtive-tFan8|X>Ft-8y8tem-i6-tlie-mâÿ€>F-Foute-leF-Gonjugate€k

bile acid uptake, whereas the passive or facilitative absorption down the length of

the small intestine may be significant for unconjugated and some glycine-

conjugated bile acids.

Being polar molecules, conjugated bile acids require transporters to

achieve efficient vectorial transport across plasma membranes. Table 1.5 lists

the transporter and carrier proteins invoived in this process. In general, the

enterocytes and the hepatocytes each have at least three members, a receptor

that binds bile acids and transports them into the ceil, a cytoplasmic bile acid

binding protein that facilitates their movement inside the cell, and an efflux

protein that pumps bile acids out of the cell. In the intestine (Figure 1.5), the

apical sodium dependent transporter (ASBT) is present on the brush border

membrane in the ileum. It recognizes and binds biie acids in the intestinal lumen

and transports them into iieai enterocytes, passing them to the ileal bile acids-

binding protein, which shuttles bile acids across the cell to the basolateral

membrane of the ileal enterocyte, where a sodium-independent efflux system

pumps bile acids to portal vein. The recently identified protein t-ASBT in rat,

which is an alternatively spliced form of ASBT, and the mutidrug resistance

protein MRP3, are suggested to be involved in this efflux mechanism [77,78].

Bile acids in the portal vein bind to albumin and flow to the liver. There they are

recognized by a transporter with high homology to ASBT, the NaMaurocholate

cotransporting golypeptide (NTCP). As illustrated in Figure 1. 6, besides NTCP,

bile acids can also be taken up into the liver by sodium-independent organic

16 - anion.transporte^proteii^(OA-TB>r^wbiclvmediatds-transport of^orgamc-anions»-

such as estrone-3-sulfate, estradiol-17 -glucuronide, as well as some

unconjugated and conjugated primary and secondary bile acids. Both NTCP and

OATP transport bile acids across the sinusoidal plasma membrane into the liver

celi where they are bound by cytosolic binding protein HBAB and shuttled to the

canalicular membrane. Conjugated bile acids are directed for immediate

secretion into bile by a ATP-dependent bile salts export pump (BSEP) located in

the canalicular membrane [79]. This protein belongs to the ATP binding cassette

(ABC) superfamily, and its activity is reguiated transcriptionally by bile acids [80].

Unconjugated bile acids are first directed to systems that conjugate them with

glycine or taurine and then secreted by the same mechanism. As one of the

anatomical components of the EHC, cholangiocytes (the cells that line the bile

ducts inside the liver) are also functionally involved in the EHC by absorbing bile

acids at their apical domain through the transporter identical to the ileal ASBT

[81,82]. Bile acids that escape hepatic clearance enter into systemic circulation,

and are either re-presented to the iiver by the hepatic artery or salvaged in the

kidney by a sodium-dependent transporter identical to ileal ASBT. These

transporters and carrier proteins allow the bile acids to be largeiy maintained within the EHC so that they can be used repeatedly in the digestion process without significant loss from the intestine to the feces or from systemic circulation to the urine.

17 ~ t.2^Sodium»dependentbUe.acid-cotransporte{S-(S6A.TsX ------

As introduced above, sodium-dependent bile acid transporters (SBATs)

are very important molecular components of the EHC in mammals, responsible

for reabsorption of bile acids in the small intestine, iiver, biie duct, and kidney.

According to their membrane location, SBATs can be grouped into apical and

basolateral SBATs, one with apical localization in ileal, kidney and in the

cholangiocytes (i.e. ASBT), and the other with basolateral localization in the

sinusoidal membrane of hepatocytes (i.e. NTCP). Apical and basolateral SBATs

have been cloned from-several species, including hamster, rat, mouse, rabbit,

and human [8388]. ASBT and NTCP share considerable amino acid identity

(32%) and very similar hydropathy plots (Figure 1.7 and 1. 8 ). It has been

suggested that ASBT and NTCP have similar membrane topologies with a

glycosylated N-terminus, seven to nine transmembrane domains and a

cytoplasmic C-terminus [57,84,89]. Detailed substrate specificity studies

revealed that NTCP has a much broader specificity than ASBT [90-92]. It was

suggested that NTCP might have evolved to a muitispecific transporter that can

aid in the hepatic clearance of steroid sulfates, organic anion metabolites and

xenobiotics [93]. This difference in substrate affinity is in agreement with the

different physiological functions for the ileum and the liver: the ileal transport

systems assimilate nutrients (including natural primary biie acids) in the manner

that emphasizes selectivity and efficiency, whereas the strict substrate

discrimination is not desired for the entry to the liver - the essential organ for

metabolism and detoxification.

18 1.2.3 Expression, regulation and transport properties of the ASBT

The human ASBT gene (SLC10A2) is localized in chromosome 13q33

[94], and encodes a 348 amino acid membrane glycoprotein. ASBT has a highly conserved sequence among different species (Figure 1.9), including human [ 86 ] hamster [85], rat [95], mouse [88 ] and rabbit (EMBL/Genebank. Accession No

254357). As the closest relative to the human ASBT, the rabbit ASBT shares

92% sequence similarity and 86 % identity with its human counterpart (Table 1.6).

ASBT is primarily expressed on the apical membrane of ileal enterocytes in mammals. In addition to the ileum, ASBT expression is shown in the renai proximal tubules [92]. The ASBT in kidney is identical to ileal ASBT. As in the ileum, ASBT in the renai proximal tubules functions as a salvage transporter to further recover biie acids that escape the enterohepatic circulation. Besides ileum and kidney, ASBT was also expressed in rat cholangiocytes (biliary epithelium)

[81,96]. Although the physiological significance of the presence of ASBT in biliary epithelium is not as distinct as that of ASBT in kidney and ileum, an interesting mechanism of the choiehepatic cycling of bile acids has been proposed [97].

How ASBT is regulated at the molecular level is stiii largeiy unknown.

Studies on familial hypertriglyceridemia (FHT) revealed that diminished gene expression of ileal ASBT in these patients results in their impaired biie acids absorption [98]. It was hypothesized that the subjects with FHT have an abnormality in the regulation of ASBT gene expression. Recently, it was found

19 mRNÆand^ protein levels and thus upregulates taurocholate uptake by the ileal brush-bcrder membrane [99], however, the mechanism remains unclear. Furthermore, there are conflicting results from animal studies on whether the iumenai biie acid levels are up- or down- regulatory factors [100-102]. Recent studies on patients with primary bile acid malabsorption revealed three naturally occurring ioss-of - function mutations in the human SLC10A2 gene [86,89]. Such dysfunctional mutations cause significant biie acid malabsorption and reduced plasma iow- density lipoprotein (LDL) cholesterol levels. These findings further underline the importance of the ASBT for maintenance of the enterohepatic circulation of bile acids and cholesterol homeostasis.

Transport studies with membrane vesicles and cell lines expressing cloned ASBT showed that ASBT - mediated taurocholate uptake is strictly Na* dependent and chloride independent and both conjugated and unconjugated biie acids can be efficiently transported by ASBT [90,92]. Electrophysiological characterization indicated that the co-transport of biie acids and Na*by ASBT is eiectrogenic and bi-directional with a 2:1 Na*: biie acid coupling stoichiometry

[103].

The famesoid X receptor (FXR), an orphan nuclear receptor, is a bile acid- activated receptor that coordinates biie acid homeostasis [104-106]. In the small intestine, biie acids enter the enterocyte through the ASBT and activate the FXR, which upregulates the iieai bile acid binding protein, a carrier protein facilitating their re-uptake by the intestine. Inside the hepatocytes, when bound to biie acids,

20 -FXR-represses-tFanseription'ef-the-gene-eneeding-eliolesterel-? -hydroxylase; which Is the rate-limiting enzyme in bile acid synthesis. In addition, the activated

FXR will further decrease bile acid uptake by reducing the level of NTCP, and stimulating the export of bile acid by increasing the expression of the bile salt export pump. Furthermore, FXR induces small heterodimer partner, an atypical nuclear receptor, which attenuates bile acid synthesis by inhibiting the action of the orphan nuclear receptor, liver receptor homolog- 1, which is a competence factor for CYP7A1 transcription. FXR hence stimulates bile acid re-uptake and controls bile acid production through a regulatory circuit involving both a nuclear receptor regulatory cascade and a number of specific transporter proteins. These findings also demonstrate a mechanism by which bile acids transcriptionally regulate their biosynthesis and enterohepatic transport (Figure 1.10).

Since bile acids are biosynthesized from cholesterol and are the only forms for cholesterol to be excreted in vivo, a specific nonabsorbable inhibitor of the ileal bile acid transporter system would lower plasma cholesterol level by blocking the intestinal reabsorption of bile acids and consequently raising the conversion of cholesterol to bile acids. Recently, several compounds have been shown to be able to lower serum cholesterol in animal studies by specifically inhibiting ASBT [107]. These findings substantiate the feasibility of using ASBT as a new pharmacological target for cholesterol-lowering therapy [75].

21 ~1^2r4-StitictoW"- actwty-relatlonship^(SAR>-f<«"ASB-T^^—

Studies with endogenous biie acids have led to physiological

understanding of the function of bile acids and the EHC. The search for a deeper

understanding of the molecular mechanism behind the affinity and recognition of

molecules by the ASBT and NTCP, has led researchers to modify bile acids and study the carrier affinity of these modified compounds. These modifications typically entail either the substitution of the hydroxyl groups at the 3 ,7 , or 12 positions by other functionalities, or the addition to or alterations at the C-17 side chain (Figure 1.4). Through the intensive studies by numerous researchers [108-

113], the following structural features essential for ASBT affinity have been revealed and generalized:

1 .The presence of at least one hydroxyl group at position 3 ,7 or 12;

2. A single negative charge in the vicinity of the C l7 side chain;

3. Substitutions at the C3 position do not interfere with active transport, suggesting C3 is a good conjugation site for the pro-drug approach.

4. Substitutions at the C-17 position do not interfere with transport as long as a negative charge around the C-24 position remains.

Accordingly, a preliminary and intuitive "pharmacophore " model of ASBT substrate binding sites was proposed to accommodate these structural requirements of the substrate:

The recognition site is a hydrophobic pocket on the membrane surface that consists of three components: a recognition site for interaction with the

22 - sterekl-nueleu8v-a-eationi&8lte-fOf-eoulornbiG^tnteraetion-witMiie'ne§atively^ -

charged side chain; and an anionic site for interaction with Na*.

Besides modifying bile acids and testing the affinity of these modified

compounds to ASBT, researchers also discovered a series of chemically diverse

ASBT inhibitors. With an increasing amount of experimental data, modeling

approaches were incorporated to further study this structure-activity relationship.

1) Using a series of 30 ASBT inhibitors and substrates, Swaan and colleagues

[68 ] constructed a 3D-QSAR model for ASBT by using comparative molecular

field analysis (CoMFA). 2). Using a training set of 17 chemically diverse inhibitors

of ASBT, Baringhaus and colleagues [69] developed a reliable enantiospecific

pharmacophore that mapped the molecular features essential for ASBT affinity:

one hydrogen bond donor, one hydrogen bond acceptor, and three hydrophobic

features. For natural bile acids they found that a) ring 0 in combination with

methyl-18 map one hydrophobic site and methyl-21 maps a second; b) a a-OH

group at position 7 or 12 constitutes a hydrogen donor; c) the negatively charged

side chain comprises the hydrogen bond acceptor; and d) the 3 a-OH group does

not necessarily map a hydrogen bond functionality. These two models are in

good agreement and should facilitate in the rational in silico design of substrates

for ASBT.

As described above, our current knowledge about the structure - activity

relationship of ASBT is mostly derived from the approaches in which the naturai

and modified bile acids and the chemically diverse ASBT inhibitors were

thoroughly studied to deduce a pharmacophore model of the ligand binding site.

23 A-fulleiLundeRlanding-o(^llwSAR^aWh6^mol@GulaFleve^Fequife&theno*-ye*- available crystal structure of ASBT. However, with the simplification of the experimental techniques in molecular biology, the dramatic improvement in computer technology and modeling algorithms, and the continuous evolution of bioinformatics, it becomes feasible to take rational steps to study the SAR by approaching the macromolecule itself, instead of solely relying on the information generalized from small substrates. First of all, the secondary structure of ASBT can be predicted from its primary amino acid sequence. To be specific, the two- dimensional (2-D) membrane orientation of ASBT can be derived from hydropathy analysis and other relevant sequence analysis tools. This putative membrane topology may then be evaluated by several biochemical and biophysical techniques. Based on the optimized 2-D topology model and the sequence alignment with an appropriate membrane protein with the available 3-D structure, a 3-D working model for ASBT can be constructed. Both 2-D and 3-D models can then be used as the working hypothesis to guide subsequent mutagenesis experiments. The results of these experiments can then be used to verify or modify the 2-D and 3-D models and thus further improve their biological relevance. By doing so, we might be able to get a reliable molecular representation of interaction sites between ASBT and its substrate even in the absence of a high-resolution crystalline structure. Thus, it can be anticipated that the combination of these approaches will provide a powerful tool for the design of novel substrates and inhibitors for this transporter.

24 —IrfcS-Use-efbile.^eid tfanspert-system-ferdfug-deHvefy------

Bile acids display a physiologic organotropism for the liver and the small

intestine, which suggests that bile acid transport pathways can be exploited at

least in two ways: to shuttle compounds across the intestinal epithelium and

target them to the liver and the hepatobiliary system. This approach was first

envisioned in 1948 by Berczeller in a US patent appiication (#2,441,129),

describing the synthesis of suiphonylcholylamide, which was claimed to be

particularly suitable for the treatment of “germ and virus diseases that attack the

liver.” It was not until the mid 1980’s when Ho [114] suggested several potentially

therapeutic applications of using the bile acid transport systems for: a) the

improvement of the oral absorption of an intrinsically, biologically active, but

poorly absorbed hydrophilic drug; b) liver site-directed delivery of a drug to bring

about high therapeutic concentrations in the diseased liver with the minimization

of general toxic reactions elsewhere in the body; and c) gallbladder-site delivery

systems of cholecystographic agents and cholesterol gallstone dissolution

accelerators. His experimental evidence was based on bile acid analogs with

minor structural modifications at the 3- position that were handled like natural bile

acids by the liver and intestine during in situ perfusion experiments. The

feasibility and viability of using the bile acid transport pathway for targeting of

drug-bile acid derivatives was more definitively demonstrated by the laboratory of

Kramer, who primarily used the 3-position for drug conjugation. Studies with

small peptides attached to the C-24 position further demonstrate the general applicability of this transport system for drug targeting and delivery purposes [73].

25 — ^From^»8 tFuetuFal-peF8 peet»veriargeiy^ba8 e€t-o(vtlie^above^eutline€^

structure-activity reiationships, the foliowing drug targeting strategies are feasibie

with bile acids:

1). Attachment of drugs to the bile acid side chain (C-17) with positional

retention of the negatively charged group;

2). Attachment of drugs to the steroid nucleus at hydroxyl positions 3, 7, or

12 with consen/ation of the C-17 side chain.

The use of carrier-mediated bile acid for drug delivery purposes can be divided into three groups: liver and gallbladder directed delivery, oral absorption enhancement, and lowering serum cholesterol.

1.2.5.1 Liver and gallbladder delivery

So far, most studies have focused on using the bile acid transport system for liver targeting. The research groups of Kramer [74,115,116] and Stephan

[117] have successfully shown specific hepatic delivery of chlorambucil, HMG-

CoA reductase inhibitors, and L-T3, respectively. These studies prove that the coupling of drug entities to bile acids does not cause a loss of affinity for the hepatic biie acid carrier. Apart from the necessity of a negatively charged group around the C-24 position, Kim and colleagues [118] have shown that some size restrictions apply when compounds are coupled to the C-3 position. Both the 3 and the 24 positions appear to be usable coupling points for a prodrug stratagem. The 24-position appears to be an attractive site in the bile acid molecule for coupling purposes. The carboxylic acid moiety is easily linked to an

2 6 "amtne-tising-eefwentfonahpepttdfr^synthesizing techniques making^the-

synthesis of these compounds relatively easy. It should be stressed, however,

that a negatively charged group around the C-24 position is required.

1.2.5.2 Systemic delivery

Using a prodrug approach with a bile acid molecule as a shuttle, liver-

targeting is easily accomplished, whereas systemic drug delivery needs to

address the problem of rapid biliary excretion. Thus far, no single study has

unequivocally shown the release of the parent compound from the conjugate

after passage across the intestinal wall. It has to be mentioned, however, that no

studies so far have attempted to develop a prodrug approach in which the drug

will be released prior to arrival in the liver. In that case, the drug moiety must be

released from the bile acid it is coupled to either within the enterocyte or the

portal vein. Only if these conditions are met, systemic delivery using a prodrug approach can be successful. Although promising, the suitability of this transport system for systemic drug delivery remains to be demonstrated.

1.2.5.3 Cholesterol-reducing agents

Hypercholesterolemia is well known as a major risk factor for coronary heart disease. In clinical practice, two main hypocholestrolemic agents are commonly used: the 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (Lipitor^), and the bile acid séquestrants, such as cholestyramine and colestipol [ 122], which bind bile acid in the intestinal lumen

27 -afKi-thtw-tncreasesbHfractcKexcwtienrThe-main-drawback^ofthesfragenty tS""-"

their non-specificity, which leads to poor patient compliance due to adverse side

effects, such as high dosages of 10-30 g per day, constipation, maldigestion and

malabsorption syndromes. As the alternative method to bile acids séquestrants,

any reagent that can specifically inhibit the bile acid active transport system could

block the reabsorption of bile acids and consequently reduce the serum

cholesterol level. So far, several molecules have been found to possess this

effect in animal studies. The first such inhibitors comprised coupling of two bile

acid molecules via a spacer to allow presumable interaction with more than one

transporter site, resulting in an efficient inhibition of bile acid reabsorption without

or with only low absorption of the inhibitor itself [123]. Recently it was shown that

a benzothiazepine derivative, 2164U90, was able to selectively inhibit active ileal

bile acid absorption in rats, mice, monkey, and humans [124,125]. Similarly,

another compound S-8921[126130], a lignan derivative, was able to reduce

serum cholesterol in hamster, mice, dog and rabbit. The inhibition of the intestinal

bile acid transporter system is thought to be the underlying mechanism for an

increased fecal bile acid excretion and lower plasma LDL cholesterol levels after

oral administration of these drugs.

1.2.6 Summary

With the cloning and functional characterization of key bile acid

transporters in the liver and intestine, we now have a better understanding of

how the different components in the EHC cooperate, the genetic disorder of bile

28 -aeid-tFansportrand-the-physiologieal-basls-fe^uee-ol-tlie-bile-aekt-transpoFt"

system for drug delivery. The further exploration of the pharmaceutical potential

of these transporter systems, such as designing novel substrates and inhibitors

for ASBT, will be greatly facilitated by a better and more detailed illustration of

the SAR of ASBT.

1.3 Significance of this dissertation work

It is predicted that 10-15% of all genes in the human genome (-30,000

genes) encode for membrane transporters. However, so far, only a small fraction

(<15%) of them have been cloned and studied in some detail. From this aspect,

transporter research is an expanding and exciting research area.

While there are many important transporters worthy of study, our

laboratory is particularly interested in the apical sodium dependent bile acid

transporter, ASBT. As with most other transporters, approaches are needed to

obtain structural information for ASBT in the absence of a crystal structure. Since

ASBT is one of the sodium-dependent transporters in the SLC transporter

superfamily, the information derived from this particular transporter will be useful

in understanding other sodium-dependent or closely related transporters.

On the other hand, ASBT itself has significant pharmaceutical relevance.

Information gained from this work will significantly increase our understanding of the structurai basis and mechanism that drives bile acid transport, which, in the long run, may aid future development of specific therapeutic strategies against hypercholesterolemia and related cardiovascular diseases.

29 MARKET MOLECULAR TARGETS %

T ransporters/Channels 30

Membrane Receptors 25

Enzymes 20

Nuclear Receptors 15

Foreign Molecules (Pathogens) 5

Table 1.1 Top 100 FDA approved drugs

30 NUMBER OF SUBFAMILY NAME TRANSPORTER ABC-A 12 ABOI ABC-B 11 MDR/PGP ABC-C 12 MRP ABC-D 4 ALD ABC-E 1 OABP ABC-F 3 GCN20 ABC-G 5 White

Table 1.2 Human ATP-binding cassette (ABC) transporter superfamily

31 Channels

UniporterS(facilitated diffusion carrier) Transporters ^ ^ Carriers—► 1® Active transporters

2® A c tiv e tra n s p o rte rs

Figure 1.1 The currently recognized primary types of human transporters

32 Trivial Cytogenetic | Symbol Name/Substrate Organ Expression Name Location ! SLC1A1 Neuronal/Epithelial High Affinity Glutamate EAAT3 Neurons, Kidney And 9p24 I Transporter, System Xag 1 EAAC1 Intestine 1 SLC1A2 Glial High Affinity Glutamate Transporter 2 GLT1, Brain 11p13-p12 ; EAAT2 SLC1A3 Glial High Affinity Glutamate Transporter 3 EAAT1 Brain, Liver, Muscle, reserved ( Ovary, Testis SLC1A4 Glutamate/Neutral Amino Acid Transporter 4 ASCT1 2p15-p13 SLC1A5 Neutral Aminp Acid Transporter 5 M7V1; 19q13.3 1 M7VS1 SLC1A6 High Affinity Aspartate/Glutamate EAAT4 CNS reserved \ Transporter 6 I SLC1A7 Glutamate Transporter 7 EAAT5 Retina reserved I SLC2A1 Facilitated Glpcose Transporter 1 GLUT1 Most Cells 1p35-p31.3 1 SLC2A2 Facilitated GIpcose Transporter 2 GLUT2 Liver, Pancreas 3q26.2-q27 SLC2A3 Facilitated Glpcose Transporter 3 GLUT3 Neurons 12p13.3 f SLC2A4 Facilitated GIpcose Transporter 4 GLUT4 Fat, Skeletal Muscle, 17p13 Ï Myocardium i SLC2A5 Facilitated GIpcose/Fructose Transporter 5 GLUTS Intestine, Brain 1p36.2 \ Microglia I SLC2A6 Facilitated GIpcose Transporter 6 GLUTS 9q34 SLC2A7 Facilitated GIpcose Transporter 7 GLUT7 Liver SLC2A8 Facilitated GIpcose Transporter 8 GLUT8 9 SLC2A9 Facilitated GIpcose Transporter 9 GLUT9 4p16-p15.3 SLC2A10 Facilitated GIpcose Transporter 10 GLUT10 20q12-13.1 SLC2A11 Facilitated GIpcose Transporter 11 GLUT11 reserved

continu^ Table 1.3 Overview of the Solute Carrier (SLC) protein genetic superfamily Table 1.3 Continued SLC3A1 Cystine, Dibasic And Neutral Amino Acid ATR1 Kidney And Intestine 2pter-q32.3 j Transporters, Activator Of Cystine, Dibasic And D2H 1 Neutral Amino Acid Transport 1 RBAT SLC3A2 Activators Of Djbasic And Neutral Amino Acid MDU1 Actively Proliferating 11q12-q22

Transport 2 Cells i. SLC4A1 Anion Exchanger, Member 1 (Erythrocyte EPB3 Erythrocyte 17q12-q21 Membrane Protein Band 3, Diego Blood AE1 Group) 1 SLC4A2 Anion Exchanger, Member 2 (Erythrocyte AE2 Leukocytes, Gl Tract 7q35-q36 | Membrane Protein Band 3-Like 1) BND3L i SLC4A3 Anion Exchangpr, Member 3 AE3 Brain Neurons, Heart 2q36 { SLC4A4 Sodium Bicarbonate Cotransporter NBC1 Pancreas, Heart, Brain 4q21 And Kidney ] SLC4A7 Sodium Bicarbonate Cotransporter7 NBC2 Retina, Testis, Spleen, 3p22 ^

Ovary, Small Intestine, 1. Colon, Thymus, Heart, SLC4A8 Sodium Bicartxpnate Cotransporter 8 NBC3 Brain And Testis Chr.12 SLC4A9 Sodium Bicarbonate Cotransporter 9 reserved SLC4A10 Sodium Bicart)onate Transporter-Like 2q23-q24 i SLC5A1 Sodium/Glucose Cotransporter 1 SGLT1 Intestine 22q13.1 j SLC5A2 Sodium/Glucose Cotransporter 2 SGLT2 Kidney 16p12-p11 ' SLC5A3 Inositol Transporters 3 SMIT Many Tissues, 21q22 Including Brain i SLC5A4 Neutral Amino Acid Transporters, System A 4 SAAT1 22q12.1-q12.9 SLC5A5 Sodium Iodide Symporter 5 NIS Primarily In Thyroid 19p13.2-19p1p Tissues, also In Breast, Colon, And Ovary SLC5A6 Sodium-Dependent Vitamin Transporter SMVT Most Tissues 2p23 SLC5A7 Choline Transporter CHT1 2q12 ; Table 1.3 Continued

SLC6A1 Neurotransmitter Transporter. GABA GAT1 Brain 3p25-p24 SLC6A2 Neurotransmitter Transporter, Noradrenalin NET Brain 16q12.2 j NAT1 SLC6A3 Neurotransmitter Transporter, Dopamine □ATI Brain 5p15.3 [ SLC6A4 Neurotransmlttpr Transporter, Serotonin SERT Brain 17q11.1-q12; SLC6A5 Neurotransmitter Transporter, Glycine GLYT1 Medulla, Spinal Cord And 11p15.1-15.q Cerebellum 1 SLC6A6 Neurotransmitter Transporter, Taurine TAUT Ileum, Brain, Liver, Heart 3p26*p24 ; And Placenta SLC6A6P Solute Carrier Family 6, Member 6 21pter-qter f Pseudogene SLC6A7 Neurotransmitter Transporter, L-Proline PROT Brain 5q31-q32 | SLC6A8 Neurotransmitter Transporter, Creatine CT1 Muscle Xq28 1 SLC6A9 Neurotransmitter Transporter, Glycine GLYT2 CNS, & peripheral tissues 1p33 SLC6A10 Neurotransmltt^r Transporter, Creatine CT2 Testis 16p11.2 j SLC6A11 Neurotransmitter Transporter, GABA GAT3 reserved SLC6A12 Neurotransmltt^r Transporter, Betalne/GABA 12p13 Î SLC6A13 Neurotransmitter Transporter, GABA 12p13.3 i SLC6A14 Neurotransmitter Transporter 14 Xq23-q24 ^ SLC6A15 Neurotransmitter Transporter 15 hv7-3 reserved SLC6A16 Neurotransmitter Transporter 16 reserved SLC7A1 Cationic Amino Acid Transporter, Y + 1 ATRC1 13q12-q14 HCAT1 SLC7A2 Cationic Amino Acid Transporter, Y+ 2 ATRC2 Intestine 8p22-p21.3 HCAT2 SLC7A3 Cationic Amino Acid Transporter, Y+ 3 CAT3 reserved SLC7A4 Cationic Amino Acid Transporter, Y+ 4 CAT4 Brain, Testis, and 22q11.2 Placenta Table 1.3 Continued

SLC7A5 Cationic Amino Acid Transporter, Y+ 5 LAT 1 Lung, Liver, Brain, 16q24.3 MPE16 Thymus, Retina, And 1 Some Other Tissues. s) SLC7A6 Cationic Amino Acid Transporter, Y+ 6 LAT3 All Tissues Tested reserved y(+)LAT2 Except Liver; Expression Was Weak : In Pancreas And ij Highest In Thymus Î SLG7A7 Cationic Amino Acid Transporter, Y+ 7 y(+)LAT1 Kidney And Intestine 14q11.2 SL07A8 Cationic Amirro Acid Transporter, Y+ 8 IA T2 Kidney 14q11.2 1 SLC7A9 Cationic Amino Acid Transporter, Y+ 9 Kidney, Liver, Small 19q13.1 1 m r Intestine, And Placenta 1 SLC7A10 Cationic Amino Acid Transporter, Y + 10 asc -1 19ql3.1 SLC7A11 Cationic Aminlo Acid Transporter. Y+ 11 xCT 4q28-qS2 I SLC8A1 Sodium/Calciÿm Exchanger 1 NCX1 Heart. Brain, Retina 2p23-p21 ! Skeletal And Smooth Muscles SLC8A2 Sodium-Cajcipm Exchanger 2 NCX2 Brain And Skeletal 14 Muscle SLC8A3 Sodium-Calcium Exchanger 3 NCX3 reserved SLC9A1 Sodium/Hydrogen Exchanger), Isoform 1 APNH Expressed Ubiquitously 1p36.1-p35 Î (Antiporter, Na+/H+, Amiloride Sensitive) NHE1 SLC9A2 Sodium/Hydrogen Exchanger, Isoform 2 NHE2 Intestine, Kidney 2q11.2 SLC9A3 Sodium/Hydrogen Exchanger, Isoform 3 NHE3 Intestine, Kidney 5p15.3 SLC9A3P Sodium/Hydrogen Exchanger, Isoform 3 NHE3P 10 Pseudogene 1 SLC9A3R1 Sodium/Hydrogen Exchanger, Isoform 3 NHERF Kidney, Small Intestine, reserved Regulatory Factor 1 Placenta, And Liver \ Table 1.3 Continued

SLC9A3R2 Sodium/Hydrpgen Exchanger, Isoform 3 NHERF-2 reserved ■ Regulatory Factor 2 SLC9A4 Sodlum/Hydrpgen Exchanger. Isoform 4 NHE4 2 SLC9A5 Sodlum/Hydrpgen Exchanger. Isoform 5 NHE5 Stomach 16q22.1 SLC9A6 Sodium/Hydrogen Exchanger. Isoform 6 NHE6 Ubiquitous X SLC10A1 Sodium/Bile Acid Cotransporter Family 1 Ntcp Liver 14 1 SLC10A2 Sodlum/Blle Acid Cotransporter Family 2 ASBT. Ileum. Bile Duct, 13q33 ! IBAT Kidney i SLC11A1 Proton-CoupljBd Divalent Metal Ion NRAMP1 Phagocytic Cells 2q35

Transporters 1 Ï SLC11A2 Proton-Couplpd Divalent Metal Ion NRAMP2. All Tissue Tested 12q13 Transporters |2 DCT1 SLC11A3 Proton-Couplpd Divalent Metal Ion FPN1. Placenta. Liver. 2q32 Transporters 3 IREG1 Spleen. And Kidney SLC12A1 Sodlum/Potasslum/Chlorlde Transporters 1 NKCC2 Kidney 15q15-q21 SLC12A2 Sodlum/Pota^slum/Chloiide Transporters 2 NKCC1 Many Tissues 5q23.3 SLC12A3 Sodium/Chloride Transporters 3 NCCT Kidney 16q13 SLC12A4 Potassium/Chloride Transporters 4 KCC1 Ubiquitous 16q22.1 ' SLC12A5 Potassium-Chloride Transporter 5 KIAA1176 20 ! SLC12A6 Potassium/Chloride Transporters 6 KCC3 Brain. Heart. Skeletal 15q13 Muscle. And Kidney SLC12A7 Potassium/Chloride Transporters 7 KCC4 Many Tissues 5p15 SLC13A1 Sodium/Sulpt)ate Symporters 1 NaSi-1 reserved SLC13A2 Sodium-Dependent Dicarboxylate NADC1 Kidney And Intestine 17p11.1-q11.1 Transporter 2 SLC14A1 Urea Transporter 1 (Kidd Blood Group) HUT11 Kidney And Red Cells 18q11-q12 ! SLC14A2 Urea Transporter 2 (Vasopression UT2. UT-A Kidney 18q12.1-q21-1 Regulated) j Table 1.3 Continued

SLC15A1 Oligopeptide Transporter 1 PepTI Intestine 13q33-q34 SLC15A2 H+/Peptide Transporter2 PepT2 Kidney resetved 1 SLC16A1 Monocarboxylic Acid Transporters 1 MCT1 Erythrocytes, Muscle. 1p13.2-p12 1 Intestine, And Kidney 1 SLC16A2 Monocart)oxy|ic Acid Transporters 2 XPCT Xq13.2 (Putative Transporter) SLC16A3 Monocarboxyfic Acid Transporters 3 MCT3 Ubiquitous reserved SLC16A4 Monocarbo)Qf)ic Acid Transporters 4 MCT4 Many Tissues reserved SLC16A5 Monocarbo}qf|ic Acid Transporters 5 MCT5 Kidney reserved SLC16A6 Monocarboxyjic Acid Transporters 6 MCT6 Many Tissues reserved SLC16A7 Monocarboxyfic Acid Transporters 7 MCT2 Erythrocytes And 12q13 ' Tumor Cells SLC17A1 Sodium Phosphate 1 NPT1 Kidney 6p23-p21.3 ; SLC17A2 Sodium Phosphate 2 NPT3 6p21.3 SLC17A3 Sodium Phosphate 3 NPT4 6p21.3 1 SLC17A4 Sodium Phosphate 4 KAIA2138 Small Intestine, Colon, 6p22>p21.3 1 Uver, And Pancreas 1 SLC17A5 Anion/Sugar Transporter 5 SIALIN Expressed Ubiquitously 6q14-q15 ( SLC18A1 Vesicular Monoamine 1 VAT1 Brain 8p21.3 1 SLC18A2 Vesicular Monoamine 2 VAT2 Brain 10q25 SLC18A3 Vesicular Acetylcholine 3 VACHT Brain 10q11.2 SLC19A1 Folate Transporter 1 FOLT Placenta, Small 21q22.3 Intestine And Other ! Tissues SLC19A2 Thiamine Transporter 2 THTR1 Expressed In Many 1q23.2 Tissues, Abundant In Skeletal And Cardiac Muscle i I -

Table 1.3 Continued

SLC20A1 Phosphate Transporter 1 GLVR1 2 SLC20A2 Phosphate Transporter 2 GLVR2 8 p12-q21 ' SLC21A1 Organic Anion Transporter 1 0AT1 reserved | SLC21A2 Prostaglandin Transporter 2 PGT Many Tissues 3q21 SLC21A3 Organic Anion Transporter 3 OATP Liver, Brain, Lung, 12p12 Kidney, And Testis SLC21A4 Organic Anion Transporter 4 OAT-K1 reserved SLC21A5 Organic Anion Transporter 5 Oatp2 reserved i SLC21A6 Organic Anion Transporter 6 LST1 Liver reserved SLC21A7 Organic Aniop Transporter 7 OatpS reserved [ SLC21A8 Organic Anion Transporter 8 OATP8 Liver reserved { SLC21A9 Organic Anion Transporter 9 OATPB Many Tissues 11q13 SLC21A10 Organic Aniop Trsmsporter 10 Oatp4 reserved SLC21A11 Organic Anion Transporter 11 OATP-D 15q26 SLC21A12 Organic Anion Transporter 12 OATP-E reserved f SLC21A13 Organic Aniop Transporter 13 OatpS reserved [ SLC21A14 Organic Anion Transporter 14 OATP-F reserved . | SLC22A1 Organic Cation Transporter 1 0CT1 Kidney, Liver, Intestine, 6q26 1 And t Colon SLC22A2 Organic Cation Transporter 2 0RCTL2 Liver And Kidney 6q26 SLC22A3 Extraneuronai Monoamine Transporter 3 0RCTL2S Gastrointestinal 6q26-q27 Tissues, Kidney And Placenta { SLC22A4 Organic Cation Transporter 4 0CTN1 Liver,Kidney, Trachea, 5 Renal And Bone Marrow Table 1.3 Continued

SLC22A5 Organic Catiop Transporter 5 0CTN2 Kidney, Skeletal 5q31 1 Muscle, Heart, And Placenta i SLC22A6 Organic Anion Transporter 6 0AT1 11q11.7 j SLC22A7 Organic Anion Transporter 7 0AT2 Liver And Kidney 6p21.2-21.1 : SLC22A8 Organic Anion Transporter 8 0AT3 11q11.7 : SLC23A1 Nucleobase Transporter SVCT2 Brain, Eye, And Other 20p13 Organs \ SLC23A2 Nucleobase Transporters 2 SVCT1 Intestine, Kidney, And 5q31.2-q31.3 | Liver SLC24A1 Sodium/Potassium/Calcium Exchanger 1 NCKX1 15q22 È SLC24A2 Sodium/Potassium/Calcium Exchanger 2 NCKX2 9p22-p13 SLC24A3 Sodium/Pota^ium/Calcium Exchanger 3 NCKX3 20p13 ! SLC24A4 Sodium/PotaœiunVCalcium Exchanger 4 NCKX4 14 Ï SLC25A1 Mitochondrial barrier; Citrate Transporter 1 CTP Liver, Testis, Ovary, 22q11 1 And Gut SLC25A3 Mitochondrial Carrier; Phosphate Carrier 3 PHC Heart, Skeletal Muscle, 12 And Pancreas SLC25A4 Mitochondrial Carrier; Adenine Nucleotide ANTI Heart And Skeletal 4q35 1 Translocator 4 Muscle SLC25A5 Mitochondrial Carrier; Adenine Nucleotide ANT2 Heart, Skeletal Muscle, Xq13-q26 Translocator 5 Liver, Kidney, Or Brain. SLC25A6 Mitochondrial Carrier; Adenine Nucleotide ANT3 Liver Xp22.32 Translocator 8 SLC25A10 Mitochondrial barrier; Dicartx>xylate Die 17q25.3 f

Transporter 1Q i. SLC25A11 Mitochondrial Carrier; Oxoglutarate Carrierl 1 OGC 17p13.3 1: Table 1.3 Continued

SLG25A12 Mitochondrial Carrier, Aralar 12 ARALAR Heart, Skeletal Muscle, 2q24 I Brain And Kidney. SLC25A13 CItrin Transporter CITRIN Liver 7q21.3 I SLC25A14 Mitochondrial Carrier, Brain 14 BMCP1 Brain, testis, pituitary Xq24 I SLC25A15 Mitochondrial Carrier; Ornithine Transporter ORNT1 Liver 13q14 ; SLC25A16 Mitochondrial Carrier; Graves Disease GDA Thyroid 10q21.3-q22.^ Autoantigen 16 SLC25A17 Mitochondrial Carrier; Peroxisomal PMP34 22q13 i Membrane Protein, 34kd 17 SLC25A18 Mitochondrial Carrier 18 reserved . SLC25A20 Camitine/Acy)camitine Translocase 20 CAC 3p21.3l SLC25A20 Camitine/Acyjcamitine Translocase 20 reserved P Pseudogene SLC25A21 Mitochondrial Oxodicarboxylate Carrier 21 14q11.2 SLC26A1 Sulfate Transrorter 1 SAT-1 4p16.3 SLC26A2 Sulfate Transporter 2 DTD Cartilage And Intestine 5q31-q34 SLC26A3 Solute Carrier Family 26, Member 3 CLD,DR 7q22-q31.1 A SLC26A4 Solute Carrier Family 26, Member 4 DFNB4,P 7q31 DS SLC26A6 Solute Carrier Family 26, Member 6 3p21 SLC26A7 Solute Carrier Family 26, Member 7 8q23 SLC26A8 6p21 SLC26A9 Solute Carrier Family 26, Member 9 1q31-q32 SLC26A10 Solute Carrier Family 26. Member 10 12ql3 SLC26A11 Solute Carrier Family 26, Member 11 17q25 SLC27A1 Fatty Acid Transporter 1 FATP Adipocytes, Skeletal reserved Muscle, Heart, Fat. Table 1.3 Continued

SLC27A2 Fatty Acid Transporter 2 FATP2 Adipose Tissue, Liver, reserved Heart, And Kidney. SLC27A3 Fatty Acid Transporter 3 FATP3 =Fatp2 reserved SLC27A4 Fatty Acid Transporter 4 FATP4 Mature Enterocytes reserved SLC27A5 Fatty Acid Transporter 5 FATP5 =Fatp2 reserved SLG27A6 Fatty Acid Transporter 6 FATP6 =Fatp2 reserved SLC28A1 Sodium-Coupled Nucleoside Transporter 1 N1.CNT1 15q25-26 SLC28A2 Sodium-Coupled Nucleoside Transporter 2 N2.CNT2 reserved - SLC29A1 Nucleoside Transporters 1 ENT1 Widely Distributed 6p21.1-p21.2 SLC29A2 Nucleoside Transporters 2 ENT2 Widely Distributed 11q13 SLC30A1 Zinc Transpojler 1 ZNT1 Ubiquitous reserved SLC30A2 Zinc Transporter 2 ZNT2 reserved SLC30A3 Zinc Transpojler 3 ZNT3 Hippocampus And reserved Cerebral Cortex. SLC30A4 Zinc Transporter 4 ZNT4 Brain, Intestine reserved SLC31A1 Copper Transporters 1 C0PT1 Most Organs 9q31-q32 SLC31A2 Copper Transporters 2 COPT2 Most Organs reserved SLC32A1 GABA Vesicular Transporter 1 VGAT reserved 1 SLC34A1 Sodium Phosphate 1 NaPi3, Kidney 5q35 I NPT2 i SLC34A2 Sodium Phosphate 2 NaPiSB Kidney, Lung, 4p15.1-p15.3 Pancreas. Prostate. SLC35A1 CMP-Sialic Acid Transporter 1 CST Ubiquitous 6 ' SLC35A2 UDP-Galactose Transporter 2 UGALT Most Tissues Xp11.23- 1 p i 1.22 SLC35A3 UDP-N-Acetylglucosamine (UDP-Glcnac) Most Tissues 1p21 Transporters SLC37A1 Glycerol-3-Phosphate Transporter 1 21q22.3 ' 1. Site directed Ala scanning 2. Membrane insertion scanning 3. Site directed thiol crosslinkers 4. Excimer fluorescence 5. Engineered divalent metal binding sites 6. Site-directed spin-labeling and electron paramagnetic resonance 7. Site-directed chemical cleavage 8. Identification of discontinuous monoclonal antibody probes 9. AAglycosylation scanning mutagenesis

Table 1.4 Techniques for studying polytopic membrane protein structure

43 V Plasma membrane

Transport vesicle

ER lumen

Rough ER

Figure 1.2 Rationale for /V-glycosylation scanning mutagenesis

AAglycosylation takes place in ER lumen. For membrane proteins, N~ glycosylation can only occur to the glycosylation sites in the lumen. The topological orientation of the membrane protein is retained during its trafficking to plasma membrane, therefore, the actualN- glycosylation site provides the topological marker for the extracelluar side of the transporter, (/sh ap es represent the glycosylation side chains)

44 Figure 1.3 Enterohepatic circulation

Bile acids are biosynthesized from cholesterol in the liver. They are stored in the gallbladder and released via the bile duct into the duodenum, where they aid in the digestion of dietary fats. Intestinal uptake of bile acids takes place along the entire length of the small intestine, but active reabsorption is confined to the distal ileum to minimize loss off bile salts in the feces. The portal circulation carries bile acids from the intestine to the liver where they are actively absorbed by hepatocytes and secreted into bile. From [131].

45 Active Transpcrt by Hepatocytu Biosynthasis by Hapatocytaa Spllovar Into Systomic Circulation Sphincter of Oddi

Storage and Emptying by Hapatic Uptaiw Gallbladder

Passive Transport by ' Jejunal Enterocytes

Filtration of Non>pfotain> bound molacules; Active Transport Active Transport by Renal Tubulas by Ileal Enterocytes

Transit to Uvar by Splanchnic and Hapatic Arterial Blood Row

Transit to Uvar by Portal Vmous Blood Row

Passive Transport by Colonic Enlerocytas

Fecal Excretion (s biosynthesis)

Figure 1.3

46 Ri R2 R3 R4 Rs Prefix -OH-OH -H-OH-OH - -OH-OH -H-OH-H Chenodeoxy- -OH-OH -H -H -OH Deoxy- -OH -OH -H -H -H Litho- -OH -OH -H -OH (P) -OH Urso- -OH -OH -H -OH (P) -H Ursodeoxy- -OH -OH (P) -H -OH (P) -H Isoursodeoxy- -OH-OH -H -H -OH (P) Lagodeoxy- -OH-OH -OH-OH-H Hyo- -NHCH2COOH - - - Glyco-1 -NH(CH2)2S03H - -- Tauro-1 ^The prefixes glyco- and tauro* prevail all others, i.e. giyco-cholic acid.

Figure 1.4 Structure and nomenclature of bile acids

47 A c c e s s io n O r g a n Na m e Lo c a t io n Fu n c t io n / C haracteristics No. (HUMAN) Intestine ASBT/IBAT Apical Sodium-dependent entry of bile acids NM_000452 BABP Cytosolic X90908 t-ASBT Basal Truncated ASBT; sodium-independent efflux of conjugated bile acids MRP3 Multiple-specific; ATP-dependent NM_003798 export of bile acids and organic anions Liver NTCP Basal/ Exclusively expressed in the liver. NM_003049 Sinusoidal Sodium-dependent entry of conjugated BA and other substrates OATP-A Multiple-specific; Sodium-independent NM_005075 OATP-C entry of bile acids and organic anions NM_006446 OATPS AJ251506 HBAB Cytosolic High affinity to bile acids AB031084 SPGP/BSE Apical/ ATP-dependent export of monovalent NM_003742 P Canaliculus bile acids Multiple-specific; ATP-dependent NM_000392 MRP2 export of organic anions including di­ anionic conjugated bile acids Bile duct ASBT Apical Sodium-dependent entry of bile acids NM_000452 Kidney ASBT Apical Sodium-dependent entry of bile acids NM_000452

OATP: Organic anion transporting polypeptides SPGR: Sister of P-gp BSEP: Bile salt export pump HBAB: human bile acid binder

Table 1.5 Bile acid transporters

48 bile acids Lumen

ENTEROCYTE

Figure t .5 Bile acid transport system ia the enterocytes

49 ' .. \

Na+ b ile a c id s Portal Vein

Bile acids 4 V biosynthesis |cyp7a

Choiestoroi

HEPATOCYTE

Figure 1.6 Bile acid transport system in the hepatocytes

50 (1)1______io______22______22______22______22______22 HUTtTLŒW 0 ) M N D P N S C #m TV C §G A S C V ^P E S N Fm iLS ^S T^TILIJU :^FS M G C N iLiE IK K FLG HurTcn_nlqD 0 ) ^ ^ Qjisa«us 0) MD A LP L LS IL LL IM SLGC ME K A

(61)61______22______i22______22______222______.110 120 HiroLCBtt (61) HiK|PW&iC^Fi|CS3|GmPLT|F|LSVAFDILlf(3A^V^|IGCCPGGTA3NILAYW^ HtrroLntap (54) ΠQnsereiB (61) HI KP GI IA L QFGIMPLTAFIL F I I AL IL I GC PGG SNI A M

(121) 121_____222______222______222______222______222____ 222 HuTai.cst»(121)GDMDLS§SMTTCSTLLALGMMPLCLLIYWMWVD-SGSIVIPYDNIGTSLVA^PVSIG HurrOLntqpO 14) G DM N LSg^TTC STFC A LG M M PLLLYIY#G IYIX]D LK D K yPYK G rVISLVL#IPC TIG QrsensiBC21)GDM LSI MTTCST ALGMMPL L lYSK D IPY I SLV LLIP SIG

(181) 181 _____222______222______212______222______.230 240 Hira\.Œtf(18Q)|FyNHKWPQKAKIl|KIGSIAGA|LIVLÏAVÿG8lLYQS--AÏÏIIAPKLWIIG^FPVAG HuTOLnfcpO 74) IV#KSKRPQYMAY0KGGMIIILÈCSVAVTVLSAmVGKSIM#AMTPLLIATSS&PFIG CbnserislJBCai) Ï L K PQ K U K G I ÎVÎVLAI FIPL S I P G

@41)241_____ 222______222______2Z2______222______.290 300 HirnarLŒttÇaS) YSLGj^LSRÎAGLPWYRCRTVÀFETGMQNTQLCSTIJÇpSFTPEEËNVVFTFPLIYSIFQ Hirncn_nlcp@34)#LIÆ#!LWFCLNGR(3WTVSMETGCQWQLCSTI%#&FPPEVIGPDFFFPLLYMIF0 GCrser6UBC241) F LGFLLÂ I L RTVA ETC QN QLCSTIxiÜiAF PE I LF FPLIY IFQ

0 0 1 )3 0 1 ______^ ______222______222______222______356 HUTtTLCBbK298) IÂF|AIFi^FYVAYKKCHGKNKAEIPESKENGTEPESSFYK!^GGEQPDEK ------Hunrcn_nkp@94 I^BGI&^IgWCYEKFKTPKDKTKMIYTAATTEETIPGAimG'I&GÊOCSPCTA QrsenSLB(301)LA À I l F Y K TE ANG F DD

Figure 1.7 Sequence alignment for human ASBT and NTCP

Created by Vector NTI Suite 6.0 Align X

51 teea

a

9# ta# 29#

Figure 1.8 Hydropathy plots for human ASBT and NTCP

52 Swiss- % % Other Amino Prot Organism Identity to Positive Name Acids PAC# HISBT to HISBT HISBT 012908 Human Ileum 348 100.0 100.0 ASBT 028727 Rabbit Ileum 348 85.7 91.7 Na-f-dependent 060414 Hamster Ileum 348 83.6 89.4 bila acid ISBT transporter ASBT 062633 Rat Ileum 348 83.0 88.2 ISBT P70172 Mouse Ileum 348 79.9 87.1 NTCP1 014973 Human Liver 349 32.0 52.2 NTCP 50.4 Na*/taurocholate Ntcp2 035940 Mouse liver 317 32.2 cotransporting Ntcpi 008705 Mouse Liver 362 31.7 49.9 polypeptide Ntcpt P26435 Rat Liver 362 31.4 49.6 Human PS proteins of P09131 477 13.8 28.0 P3 Placenta unknown Mouse genome P21129 182 10.6 22.1 functions P3 (fragment)

Table 1.6 Sodium-dependent bile acid transporters

53 0 ) 1______i2______^0______20______2°______52______52______Z2 Himn_CBa 0) -MNDPNSCVDNATVCSGASCWPESNFNNILSVVlSTVLTILLALVMFSMGCNVEIKKFLGHEKRPîVGIC Han*ar_Œtt 0) -MI3NSS|CNPNAT|CEC3DSCIAPESNFNAILSVVMSTVLTILLALVMFSMGCNVE^HKFLGHE|iRPWGIV MuCLCsa 0) -MIMSS§CPPNATVCEGDSciA^ESNFNAfCiNTVMSSVI-TTLLAMVMFSMGCNV0|HKFLGHÎKRPWGIF Rabt_aB 0) MSNLiVGCIANATVCEC»SCWAPESNEl!»ILSVVSSTVLTIUALVMFSMGCNVEiEKKFLGHIRRPWGIF R d _ a a 0) -MDNSSgCSPNATFCEGDSCiVTESNFNAILSTVMSTVLTILLAMVMFSMGCNVEINKFLGHIKRPWGIF CCTSersus 0) MDNSSVC PNATVCEGDSCWPESNFNAILSWMSTVLTILLALVMFSMGOJVEI KFLGHIKRPWGIF

(71) 71______go ______22______522______252______522______522______H 2 HuTOn_att (70) vgflcqfgimpltgfilsvapdxlpêqavwlx I gccpggtasnilaywvdgdmdlsvsmttcstllalg H orrjlaiatt (7Q) VGFLCQFGIMPLTGF§LSVAFGILPVQAWVLI^CCPGGTASNILAYWVDGDMDLSVSMTTCSTliALG MM€LŒtt (70) VGPLCQFGIMPLTGFILSVASGILPVQAWVLIMGCCPGGT§SNILAYV«DGDMCLSVSMTTCSTUJ\LG Rcdt_CBtt (71) IGPLCQFGIMPLTGFSL|VAFGliP|QAVVVLIMGCCPGGTASNILAYWVIX3DMDLSVSMTTCSTLLALG R d _ a a (70) w f ]xqfgimpltgfiiævasgi £ pvqavvvlimgccpggt | snilaywidgdmdlsvsmttcstllalg ccrsensus (71) vgflcqfgimpltgfilsvafgilpvqavwlimgccpggtasnilaywvtcdmdlsvsmttcstllalg

041) 141______250______222______.170 .180 ______222______.200 210 Hircn.att040)MMPLCLLiyrKMWVDSGglVIPYDNIGTSLVALVgPVSIGMFVNBIKWPQKAKIILKIGSIAGAILIVLIA HOT»ler_aa04Q) MMPLCLFXYTKMWVDSGTIVIPYDSIGTSLVALVIPVSIGM|tVNHKWPQKAKIILKIGSIAGAILIVLIA Mxee.cstt04® MMPLCLFÜYTKMWVDSGTIVIPYDSIGISLVALVIPVSFGMFVNHKWPQKAKIILKIGSITGVILIVLIA R(±it_aa041) MMPLCL^YTKMWVDSGTIVIPyDNIGTSLVALVgPVSIGMFVNHKWPQKAKIILKVGSIAGA^IVLIA Rd_CBa040)MMPLCljFIYTKMWVDSGTIVlPYt)SIGISLVALVIPVSIGMFVNHKWPQKAKIILld:GSIAGAILIVLIA OrsensuB 041) MMPLCLFIYTKMWVDSGTIVIPYDSIGTSLVALVIPVSIGMFVNHKWPQKAKIILKIGSIAGAILIVLIA

(211) 211______220______g3o______240______222______522______.270 280 HuncrLCBaClCD WGGILYQSAWIIAPKLWI IGTIFP^GYSLGFLLARIAGLPWYSCRTVAFETGMQNTQLCSTIVQLSFT HcnilB_CsaC210) WGGILYQSAWTIEPKLWIIGTI3reÏAGYGLGFFUVRIAGQPWYRCRTVALET(aÈQNTQLCSTIVQr,SFS MaBe_aB(210) VÎGGILYQSAWIIEPKLWIIGTIFPIAGYSLGFFLARÈAGQPWYRCRTVALETGMQNTQLCSTIVQLSFS Rcat_CStt(211) VVGGILYQSAWIIEPKLWIIGTIFP^GYSLGFFLARiAGQPWYRCRTVALETGMQNTQLCSTIVQr,SFS RcLatt(210) WGGILYQSAWIIEPKLWIIGTIFPIAGYSLGPFLARLAGQPWYRCRTVALETGMQNTQLCSTIVQLSFS OrsersuBgll) WGGILYQSAWIIEPKLWIIGTIFPIAGYSLGFFLARIAGQPWYRCRTVALETGMQNTQLCSTIVQLSFS

Ç81) 281______222______222______252______522______552______349 Hunn.ŒBC80) PEBLNVVFTFPLIYSIFQLAFAAIFLGFYVAYKKCHGKNKAEIPESKroGTEPESSFYKANGGFQPDEK H orm .attC aO ) PEDIltt.VFTFPLIYSIF(:^AFAAICLGAYVAYKKCHGKNNTELQEKTDNEMEPRSSFQETNKGFQPDEK MlBe_aaCaO) P E D im VFTFPLIYTOiFQLVFAA^SiLGÿYVTYRKCXGKMDAEFLEKTDNiMipSRPSFDETNKGFQPDEK Rcbt_attC81) PËDLTYVFTFPLIYSrF^AFAAIFXÆIYVAYSKCHGKNnAEFPBIKDTKTEPESSFHQMNGGFQpi-- RdLoaCaCD PEDLOT,VFTFPLIYfgFQLVFAAl|LXaiYVTYlbvrHGKNDAEFliKTDNBMijPMPSFQETNKGFQPDEK OrsensuBCSl) PEDLNLVFTFPLIYSIFQLAFAAIILGIYVAYKKCHGKNDAEF EKTDNEMEP SSF ETNKGFQPDEK

Figure 1.9 Multiple sequence alignment of ASBTs

Created by Vector NTI Suite 6.0 Align-X

54 Figure 1.10 Model of transcriptional control of bile acid and cholesterol

homeostasis by nuclear receptors

LXRa is a liver X receptor, which activates bile acids synthesis after bound to oxysterols. Adapted from [106]

55 Cholesterol (liver)

Oxysterols Synthesis

FXR

Bile Adds (liver)

enterohepatic FXR Transport circulation

Bile Adds (intestine)

Excretion

Figure 1.10

56 CHAPTER 2

MEMBRANE TOPOLOGY OF ASBT

2.1 Introduction

The enterohepatic circulation efficiently conserves the human bile acid pool. Vectorial transport of conjugated bile acids in hepatic and intestinal membranes is achieved efficiently by two related sodium dependent bile acid transporters (SBATs). The intestinal transporter, designated ASBT (apical sodium dependent bile acid transporter, SLC10A2) was cloned from human, hamster, rabbit, mouse and rat [85,86,88,95]; it is located on the apical domain of cells lining the ileum and reabsorbs bile acids from the intestinal lumen into lleocytes. ASBT was also detected in cholangiocytes where it was believed to provide a cholehepatic shunt pathway for conjugated bile acids that may help regulate bile formation by both hepatocytes and cholangiocytes [81]. ASBT is a

41 kDa, 348 amino acids glycoprotein; dimers of approximately 90 kDa may also occur [132]. The liver Na^-dependent taurocholate-cotransporting polypeptide, designated Ntcp (SLC10A1) [83,84], is localized selectively on the basolateral

(sinusoidal) domain of hepatocytes, and is the protein principally responsible for

57 -bile *acid II ------transport------. . - r. from portal Br's. ■■■■■!blood I I MU» Into ■ Mil ll the II II ■ hepatocyte I I II I IJII I ■ ■■ 1 T [1331. ^ ASBT ^— and- - ■ - ^ Ntcp -- ■

share 32% sequence Identity, over 52% sequence similarity, an almost

overlapping hydropathy plot (Figure 1.8), and possibly similar transmembrane

topologies [132].

In the absence of a high-resolutlon structure, there are numerous

approaches to study the topography of polytopic integral membrane proteins,

such as in vitro translation of constructs containing transmembrane sequences

[56,134], N-glycosylatlon scanning mutagenesis [13,17,44], and epitope

localization [19]. Although each Individual method has Its merit, a single method

rarely facilitates definitive topographic analysis of a polytopic integral membrane

protein.

Hydropathy analysis of SBATs predicts seven to nine putative transmembrane (TM) regions. Previous studies confirmed an exoplasmic N- termlnus and a cytoplasmic C- terminus (Nexc/Ccyt) for Ntcp [84], suggesting an odd number of transmembrane (TM) segments. In fact, most early reports adopted a 7TM model. Recently, however, Hallén and colleagues [57] suggested the possibility of a 9TM model based on membrane Insertion scanning analysis.

This topology predicts two membrane spanning regions of only 9-10 amino acids each (TM 3 and 4) and four TM regions (TM2-TM3 and TM8-TM9) that are linked by two and four extramembranous amino acids, respectively. The amino acid length of TM3 and TM4 would Indicate a p-sheet conformation for these segments to span the membrane; Interestingly, a combination of a-hellces and p-

58 sheet TM segments is uncommon for polytoplc membrane proteins and

prompted us to investigate this apparent controversy (Figure 2.1).

To differentiate between a seven and a nine TM model for ASBT, we

report an AAglycosyiation scanning mutagenesis approach based on the divergent localization of specific protein segments within the two hypothetical models (Figure 2.2). Since /V-glycosyiation occurs only on the luminal side of the endoplasmic reticulum, this method was used successfully to localize extracellular domains of receptors, channels and transporters [13,17,44]. After identifying one endogenous /V-giycosylation site (Gln^°) in wild-type human

ASBT, we engineered the AAglycosylation consensus sequence (NXS/T) into a biologically functional aglyco-ASBT (N10D mutant). Our data showed that the newly introduced consensus sequences at residues 113-118 and 266-272 could be glycosylated, which strongly supported a seven transmembrane helical model of ASBT.

2.2 Experimental procedures

2.2.1 Materials

[^H] Taurocholate acid (50 Ci/mmol) was purchased from American

Radiolabeled Chemicals, Inc (St. Louis, MO). Suifo-NHS-LC-biotin was purchased from Pierce (Rockford, IL). Unlabeled taurocholic acid and other reagents, unless noted otherwise, were from Sigma (St. Louis, MO). Restriction enzymes were obtained from Life Technologies, inc (Rockville, MD), or Roche

Molecular Biochemicals (Indianapolis, IN).

59 2.2.2 Cell culture

The monkey kidney fibroblast cell line COS-1 was obtained from the

American Type Culture Collection (CRL-1650). Cells were grown in Duibecco's

modified Eagle’s medium containing 10% fetai caif serum, 4.5g/L glucose, 100

units/ml penicillin, and 100 pg/ml streptomycin (Life Technologies, inc.) at 37 °C

in a humidified atmosphere with 5% CO2.

2.2.3 Antibody preparation

Peptide synthesis and antisera preparation were performed by Sigma-

Genosys (The Woodlands, TX). Polyclonal antiserum was raised against a peptide epitope corresponding to carboxylic terminus (amino acids 335-348) of the human ASBT (SFQETNKGFQPDEK). One cysteine residue was added to the amino terminus to allow conjugation to the carrier protein keyhole limpet hemocyanin (KLH). Two New Zealand rabbits were immunized with 200pg/rabbit of the conjugated peptide in complete Freund's Adjuvant, followed with two immunizations by 100 pg/rabbit of the conjugated peptide in incomplete Freund’s

Adjuvant. The IgG was purified from rabbit serum by Protein A-agarose (Life

Technologies, Inc.) affinity chromatography and used for westem blotting.

2.2.4 Site-directed mutagenesis

Site-directed mutations were introduced using a QuickChange kit from

Stratagene (La Jolla, CA) according to the manufacturer’s protocols. The pCMVS vector containing human ASBT cDNA was a kind gift of Dr. Paul Dawson, Wake

6 0 Forest University (Winston-Salem, NC) and was used as the template In the PCR

mutation reactions. Mutagenesis primers used to remove and insert theN-

glycosylatlon sites were designed using the Vector Suite6 software from InfoMax

(Bethesda, MO) and synthesized by Operon (Alameda, CA). Plasmids were

purified using a purification kit from Roche (Indianapolis, IN). All mutations

(Figure 2.3) were verified by DMA sequencing using a 3700 DMA analyzer

(Applied Biosystems, Foster City, CA) at the Plant-Mlcrobe Genomics Facility of the Ohio State University.

The endogenous AAglycosylatlon site (Gln'°) of human wild-type ASBT was eliminated by the mutation N10D to simplify Interpretation. The resultant mutant was used as a mutagenesis template for the Introduction of novelN- glycosylatlon sites. As shown In Figure 2.2, the nested AAglycosylatlon consensus sequence (NXS/T) was engineered Into strategic locations of the aglyco-ASBT (N10D).

2.2.5 Transient transfection of ASBT In COS-1 cell

Transient transfection was conducted using the LIpofectAMINE PLUS^ reagent (Invltrogen, Carlsbad, CA). Mock transfection with pCMVp was used as the control. Optimization of transfection was achieved by testing a series of conditions in which the amount of LipofectAMINE^** Reagent used for each well was fixed while the amount of both DNA and PLUS^^ reagent was varied. A p- galactosldase assay kit (Invitrogen, Carlsbad, CA) was used to compare the transfection efficiency under each tested condition by determining p-

61 galactosidase-rw^TT* T_= .f,- ' activity iia-tr in COS-1 ^ - n i-i ■ i cells transfected i i bv i i>ii ■ d CMVB. .. - *-rggB— —The--- r t terwoi. levels e.-t - - —a ^ - .rof . ' tA". active A . .

P-galactosldase expressed In transfected cells were analyzed by the enzymatic

assay using ortho-nltrophenyl-beta-D-galactopyranoslde (ONPG) as the

substrate according to the manufacturer's protocols. In addition, a g-gal staining

kit (Invltrogen, Carlsbad, CA), which allows visualization of the transfected cells

expressing lacZ, was used to determine the transfection efficiency In pCMVp-

transfected COS-1 cells under each condition. These two assays gave consistent

results and suggested an optimum transfection condition. In parallel experiments,

the magnitude of [^H] TCA uptake in COS-1 cells transfected by pCMVSASBT

under this optimum condition was found Indeed highest among all the conditions

tested. This optimum condition, as described as follows, was then used In all

subsequent experiments.

On the day before transfection, trypslnlzed COS-1 cells were suspended

In antlblotlc-free medium and seeded in 24-well plates at a density of 6 x 10* cell

per well. On day 1,0.4 pg/well of wild type or mutant plasmid was diluted by 21 pi

of D-MEM, mixed with 4 pi of PLUS Reagent, and Incubated at room temperature

for 15 minutes; then the pre-mlxed ONA-PLUS reagent solution (25 pi) was

mixed with the 25 pi of the diluted LIpofectAMINE^'^ reagent (2.0 pi In 23 pi of D-

MEM) and incubated at room temperature for 15 minutes; while the complex was

forming, the culture medium on the cell was replaced by 200 pi of D-MEM containing 0.1 mM non-essential amino acids but no serum or antibiotics; 50 pi of the ONA-PLUS- LIpofectAMINE^ reagent complex was added to each well, and

62 incubated at 37 ° C at 5% CQaLAfter S-àQuna incubation, the medlum vvas ____

replaced with complete medium, or with complete medium containing 8 pg/ml of

tunicamycin. On day 3 to 4 (48 -7 2 hr after transfection), the transfected COS-1

cells were processed for the TCA uptake assay, western blot analysis and cell

surface biotinylation.

2.2.6 Uptake assay, western blot and cell surface biotinylation

All of these studies were conducted largely as described previously [86].

Briefly, transfected COS-1 cells in 24-well plates were incubated in 0.2 ml of

uptake medium with the indicated amount of [‘^H] TCA for 10 min. After the washing steps, the cells were lysed by 0.3 ml/well of PBS with 1% TritonX-100, then the aliquots were subjected to both liquid scintillation counting and protein quantification by BCA protein assay kit (Pierce, Rocklord, IL). Uptake activity was determined as the number of pmols of [^H]TCA taken up per min and per mg of protein.

For immunoblotting studies, the transfected COS-1 cells in the 6-well plate were washed by PBS and lysed in 0.5 ml of lysing buffer as described [86].

Lysates were separated by a 12.5% SDS-polyacrylamide gel, separated proteins were transferred onto Immun-Blot PVDF membrane by semi-dry immunoblotting apparatus (Bio-Rad, Hercules, CA). The blot was probed with rabbit anti-ASBT antibody (1:1000 dilution) and visualized using a biotinylated anti-rabbit IgG and a chromogenic detection system (Westem blotting kit form Vector Lab,

Burlingame, VA).

63 Foc>C8ll-surfac&.biotinylatioOr^the.transfectecl-CO&*Vcells-intiifr&-well

plates were washed by PBS (pH 8.0) and Incubated In PBS containing O.Smg/ml of Sulfo-NHS-LC-BlotIn for 30 min at room temperature. After the washing steps, the cells were lysed and centrifuged. The supernatant was subjected to

Immunoprécipitation using the antl-ASBT antibody and protein A-agarose. After electrophoresis and electroblotting, the final blot was blocked with 1x casein solution and detected with streptavldln-conjugated alkaline phosphatase and chromogenic substrates (Vector Labs, Burlingame, CA).

2.2.7 Data Analysis

Nonlinear regression fits of experimental and calculated data were performed with SIgmaPlot, which uses a Quasl-Newtonlan nonlinear least squares curve-fitting algorithm. The statistical analysis was done with data from single experiments. All the experiments were repeated a total of 2-4 times and In all cases gave essentially the same results. Data with error bars represent the mean ± S.D. for triplicate samples.

2.3 Results

2.3.1 Transient transfection

Optimal transfection conditions were achieved at 0.4 pg plasmid, 2.0 pi

LIpofectAMINE^** and 4.0 pi PLUS^** reagent per well In the 24-well plate. The activity of the expressed galactosidase In three Independent transfections was

64 ~6.5^2.6romoi/lir/mg.pfoteinrancl-about35>45%-of-cell8-in-7-&Fan€lem^

fields of view) were stained in blue, suggesting a 35-45% transfection efficiency

can be reached in COS-1 cells by the LIpofectAMINE PLUS^*^ reagent.

2.3.2 AAlinked glycosylatlon in ASBT

Human ASBT has two consensus AAglycosylation sites, Asn'° and Asn^°,

which are localized on N-terminus and C-terminus of ASBT, respectively. Using

site-directed mutagenesis, we have disrupted these two sites by converting the

Asn residues of each of the NXTsequences into Asp (Figure 2.3 A), and

prepared two single mutants (N10D and N328D) and one double mutant

(N10/328D). Mutant and wild-type ASBT were transiently transfected in COS-1

cells, and the transport activity and the molecular weight of each mutant protein

were assessed and compared to that of wild-type. The wild-type ASBT (Figure

2.4A lane 2) is detected in two bands (indicated by the armws), the

unglycosylated form running as a band of 38 KDa, and the 41KDa glycosylated form, which can be completely removed after the TNM treatment {lane 3). Cell surface biotinylation confirmed membrane expression of these mutants, suggesting that both glycosylated and non-glycosylated forms of ASBT are expressed on the cell surface. The specificity of the antibody is verified by the fact that no band is detected when the cells are transfected with the mock vector pCMVp (lane 1). The mobility of N10D (lane 4 and 5) is identical to that of wild- type treated with tunicamycin, whereas N328D migrates in the exact same way as the wild- type with or without treatment by TNM (lane 6 and 7). In addition, the

65 „mobility-otthe.double.inutantMtOZ328D.(ianeL8)-is.identical tathal ottitOD^

These results confirmed that Asn^° Is the only AAglycosylation site in human

ASBT, and is consistent with the results from the previous study in which

endoglycosldase Hf and peptide: N-giycosidase F were used to remove the

glycosylated moiety in ASBT and suggested that there is only one N-

glycosylation site in ASBT [86]. In fact, absence of glycosylatlon at Asn^^° is in

agreement with the cytoplasmic location of ASBT C-terminus. In addition to the

41 and SBKda bands, bands with lower molecular weights were also present in

some cell extracts. Since these bands were not shown in the mock-transfected

cells, we suspect that they are derived from the proteolytic fragments of ASBT. In

fact, the similar bands were also observed previously when an antibody directed

against an epitope in hamster ASBT C-terminus was used [86].

The single mutant N10D retained 48% of wild type TCA uptake activity

(Figure 2.5). The activity of the double mutant N10/328D was comparable to that

of N10D, whereas the single mutant N328D was comparable to the wild type.

Taken together, these data suggest 1) that Asn^° is the only /V-glycosylation site

in ASBT and the removal of the sugar moiety may slightly influence the transport

function, and 2) Asn^° is not a /V-glycosylation site, nor is it likely located in the functional important region of ASBT. Further kinetics studies showed that the calculated Kt for TCA was approximately identical between wild-type and mutant

N10D (11.3 ± 1.9 and 12.8 ±2.1 pM, respectively), suggesting that N10D retains the similar bile acids affinity; however, a 50% reduction in the Jmax values was observed (wt: 312.5± 12.8; N10D; 156.5 ± 6.7 pmol.min'\mg protein'^).

66 T&detennine-wlietlief the-impaired-uptakd-e^mutant&wa&due'to-a decreased expression or to a deficient trafficking of the protein to the plasma membrane, western blotting was performed on both whole cell lysates and immunoprecipitated samples after cell surface biotinylation (Figure 2.4B).

Densitometric analysis suggests that the band intensity of N10D was roughly equal to the combined intensity of the two bands of the wild type ASBT on both immunoblots, suggesting that the degiycosylation may not significantly change both the total and membrane expression of ASBT. Although surface expression of wild-type and mutant ASBT is similar, their Jmax is significantly different. One explanation is that the Jmax values only reflect the apparent uptake capacity of the transporter. It depends not only on the transporter expression level on cell surface, which can be influenced by different transfection efficiency in independent experiments, but also on the rate constants for the inward translocation of the complex and outward translocation of the free transporter

[135]. Therefore, it is possible that the obsen/ed differences in Jmax values could be caused by changes in the translocation rate constants rather than the cell surface transporter density. Regardless, the aglco-ASBT (N10D) is biologically functional overall, which means that the N10D mutant retains the same conformation as wild-type ASBT, and could serve as the template for introducing consensus glycosylation sites in the putative hydrophiiic loops of the transporter.

67 . 2^.3.Mefnbrane-topology- o tASBX.

Figure 2.2 Illustrates the location of the Inserted AAglycosylatlon sites In

relation to the hypothetical 7TM and 9TM models. In the 7TM model, the inserted

consensus AAglycosylatlon sites are In extracelluar loop (EL) 1 (N^^^ISNYS^^and

N’^^LS) and EL3 (N^®®TSLNST*^^) and could be accessible for glycosylatlon; In

the 9TM model, however, these sites would be present In either the Intracellular

loop (IL)1 and TM8 and not accessible by glycosylatlon machinery.

Figure 2.6 and Figure 2.7 document the glycosylatlon status and uptake

activity of the mutants with the engineered glycosylatlon sites. In the absence of

tunicamycin, EL1 mutant presents a high molecular weight band (as indicated by

the am>v^ Identical to the mobility of the glycosylated ASBT. This band disappears In the presence of tunicamycin, suggesting that either Asn'^^ or

Asn^^® In this mutant Is Indeed glycosylated, although the glycosylatlon In this position does not appear efficient, as evidenced by the relatively faint band of the glycosylated mutant. Similar results are observed In the EL3 mutant: It seems that only one engineered site on either Asn^^ or Asn^^° Is being used for the anchor for ollgosaccharyl chain and the glycosylatlon occurs more efficiently. The fact that both 113-116 (ELI ) and 266-270 (ELS) regions are capable of incorporating a sugar moiety confirms their extracelluar location as predicted by the 7TM model.

In the presence or absence of tunicamycin, the protein band with the engineered glycosylatlon site on Asn^^^ (D124N) has the same mobility as that of the N10D mutant, suggesting that no glycosylatlon occurs on this position.

68 “-Accofding-tathfr TTM-modet position t24-is~atso locaH 2ed~extracettutarly~an(f ^

could be subjected to glycosylatlon. It should be pointed out that a negative result

from this mutation does not necessarily reject our hypothesis. Depending on the

length of the extracellular loop and steric location of the inserted glycosylation

site, not all of the sites on the extracellular loops are capable of yielding

glycosylation using this approach. Our explanation is that the location of Asn^^^ is

just six amino acids away from the putative membrane border, making it too

close to the membrane to be accessible by ollgosaccharyl transferase. In fact, it

was reported previously that a distance of at least 10-12 amino acid residues

between the inserted site and the membrane boundary is necessary for efficient

glycosylation [136].

Both ELI and EL3 mutants have no uptake activity as shown in Figure

2 7 . The double mutant N10 0 /0 1 24N also has a significantly diminished activity.

Westem blot and cell surface biotinylation studies showed the expression level of each mutant was no significant changed in compared with those of wild-type and

N100, and these mutants were expressed on the membrane. These data suggested that both extracellular loopi and 3 were likely to be in the regions essential for ASBT function.

2.4 Discussion

The elucidation of the two-dimensional orientation of transmembrane spanning region is the prerequisite step toward identifying the sodium and substrate interaction sites in ASBT and studying the translocation mechanism.

69 . . ———Pfevieti8-8H»

shown that /V-glycosylation occurs on the N-terminal (Asn^ and Asn'^) of rat Ntcp

[84]. In addition, immunostaining on primary rat hepatocytes showed that the C-

terminai of Ntcp is only subjective to antibody-antigen interaction when the ceil

membrane was permeabilized by the detergent. The topology of the trans­

location of C- and N-termini of NTCP thus derived has been extrapolated into

other transporters in this transporter family such as ASBT. Our data (Figure 2.4)

further confirm this topology by showing that only Asn^° in ASBT N-terminal is

used for glycosylation while another potential site (Asn^°) in C-terminal is not.

The substitution of Asn to Asp (N328D) has little effect on the uptake

activity (Figure 2.5). This result agreed with the conclusion that ASBT C-terminal

is not essential for transport function, due to its cytoplasmic location as well as

being the least conserved region in this transport family. In fact, the cytoplasmic

tails of both ASBT and Ntcp have been proposed important for plasma

membrane targeting of the transporter [137,138].

In vitro translation analysis [57] was used to study the membrane topology

of ASBT. This approach can be used to identify the putative transmembrane

segment and was proven useful in providing valuable topology information for

some transporters. However, it failed to definitely differentiate between the 7TM

and 9TM models of ASBT, due to the limitation of the approach itself and the

subtle difference of the TM regions between 7TM and 9TM models. In fact,

recent papers [139,140] still interpreted the experimental data according to both

models without favoring one over the other. In this study, we performed an /V-

70 -glyeo8ylatiofv«eanning-mutagene8i8-to-differantiatd-betweerv7TM^ancU9-TM>

model for ASBT. W e Inserted the AAglycosylation consensus sequences into the

regions with conflicting location prediction by two models and examined their

accessibility to the glycosylation machinery. We used the functional aglyco-ASBT

(N10D) as the template so that AAglycosylation would only occur at the inserted

sites, and we increased our chance of getting positive result by inserting two

consecutive consensus sequences in some cases, e.g. ELI and EL2.

The ability to incorporate glycosylation is essentially dependent on the

location of the acceptor site. It has been generalized that /V-glycosyiation rarely

occurs if the hydrophilic loop is too short (less than 30 residues) or the acceptor

site is too close to the membrane (less than 10 residues). Having these principles

in mind, we focused on the ELI and ELS loops predicted by the 7TM model, both

of which are predicted to be relatively long (31 and 33 residues, respectively),

and engineered the glycosylation acceptor sites in the middle of the loops by the

simple substitution of the appropriate residues. Our data showed unambiguously

that the two putative hydrophilic loops ELI and EL3 are localized extracellularly.

As a matter of fact, the other short loops could have been targeted too, such as

the EL2 in the 7TM model, and the other EL and CL loops in the 9TM model.

However, in these cases, the loops need to be elongated for insertion of the consensus sites to meet the length threshold for the potential glycosylation to occur. In addition, a larger disturbance to the intrinsic loop structure might be expected after the loop elongation, therefore, a cautious data interpretation may be anticipated as well.

71 . . . . . „^ln.generalrt^ydropathy-anaiysis-prevides-u&nol^cmiyIhe^numbeF-ef^the-

membrane spanning regions, but also the putative membrane boundaries. To

test for the exact location of membrane border Is a difficult task In the absence a

3-D structure. However, since the active site of ollgosaccharyl transferase has a

sharp distance threshold above the membrane, AAglycosylatlon scanning may be

able to provide some useful Information. For example. In the case of the double

mutant N10D/D124N, no glycosylatlon could be detected. This can be Interpreted

as evidence that the predicted location of Asp124 Is too close to the membrane

to be accessible by the glycosylatlon machinery. Theoretically, If we Intensively

probe the region near Asp124 with engineered glycosylatlon sites, we may be

able to have a better Idea of how close a particular residue Is to the membrane.

Instead of solely relying on the hydropathy plot.

It Is Interesting to note that the two mutants (ELI and ELS) are silent In

transport function although the glycosylatlons do occur In the Inserted sites and

both non-glycosylated and glycosylated mutant proteins are expressed on the

cell surface (Figure 2.6 and 2.7). These results suggest that the double mutant

LI 15S/A116N (EL1 mutant) and Q268S/C270N (ELS mutant) cause the loss of

function, and therefore, that either Leu^^^ or Ala^^® or both residues are likely

Important for substrate Interactions, and the same statement could also be true

for Gln^“ and Cys^°. In fact, localization of C y s ^ In a sodlum-sensltlve region of

ASBT has been suggested [139]. In addition, the extracellular loops 1 and S with

their underlying transmembrane regions are the most conserved regions In the

sub-family of bile acid transporters. These consented regions may be Involved In

72 “the‘bincKng-and-transpoitoFsoditfm-ancl“bHe-aeids-f14^t)rS»far, three-misaensfr

and loss-of function mutations L243P, T262M and P290S have been uncovered

in patients with primary bile acid malabsorption or Crohn's disease [86,89].

These three residues are located in either extracellular loop 3 or the underlying

membrane-spanning region. Therefore, it becomes clear that a more extensive

site-directed mutagenesis study that targets these highly conserved regions may

be desired to narrow down the sodium and substrate interaction sites.

In conclusion, our results strongly support the 7TM membrane topology

model for ASBT. These findings provide a starting point to rationalize the

experimental design to elucidate the structural basis of the transport mechanism.

In addition, this model may be relevant not only for ASBT, but also for other

transporters in the bile acid transporter subfamily, such as Ntcp.

73 Figure 2.1 The 7TM versus 9TM membrane topology model for ASBT

The predicted topological location of the ioops connecting two transmembrane segments Is largely different between these two models. For example the extracellular loop 1 predicted by 7TM model is located intraceilularly according to the 9TM model. The extracellular loop 3 predicted by 7TM model is primarily in the transmembrane domain according to the 9TM model.

74 Extracellular

NT ELI EL3

13027 180 191 253 287

TM1 TM2 TM3 TM5 TM6 TM7 7TM

150 160 215 310227

IL2 IL3 IL1

EL2

ELI 89 94 13! 164 216 225H 278 290

TM1 TM2 TM 3 TM4 TM 5 TM6 TM7 TM8 TM9 9TM

196

IL3 ilV - ^

Intracellular

Figure 2.1

75 Figure 2.2 The strategic locations of the Inserted new Af-glycosylatlon sites

Three N-glycosylatlon sites (A, B and C) were engineered Into ASBT to differentiate between the 7 and 9 TM models. According to the 7TM model, the

Inserted sites are In either the extracellular loop 1 (either In the middle of the loop or a site close to the membrane boundary) or loopS, and could be accessible by the glycosylatlon machinery; according to the 9TM model, however, these sites would be present In Intracellular loop2 or transmembrane region and would not be glycosylated.

76 Extracellular

EL2

Extracellular NT y EL2

ELI

E ■-■■■I

IL2 IL1

Figure 2.2

77 Figure 2.3 Sequences of the mutagenesis primers

(A) Primers for the removal of the endogenous /V-giycosylation sites. The numbers indicate the endogenous /V-glycosylation sites. (B) Primers for the introduction of the strategic /V-glycosyiation consensus sites. The numbers indicate the inserted potential N-glycosylation sites. Italics indicate the mutations.

EL1, the engineered sites located in extracellular loopi between residue 113 and

116; EL1, the engineered sites located in extracellular loop3 between residue

266 and 270.

78 A

WT PN SC VD N ’°ATVC ccg aac age tgt gtg gac aat gca aca gtt tg

N10D P N S CV DO" A TVC ccg aac age tgt gtg gac gat gca aca gtt tg

WT P E S K E N “*GTEPE cca gag age aaa gaa aat gga acg gag cca gag N328D P E S K E O®" G T E P E cca gag age aaa gaa gac gga acg gag cca gag

B N10D G T A S N"M L A Y W V D G gga act gcc tcc aat ate ttg gcc tat tgg gtc gat ggc ELI G T A S N"N 5 A/'"Y S V D G gga act gcc tcc aat ate tcc aac tat teg gtc gat ggc N10D GMQN“*TQLCST IVQ ggg atg cag aac acg cag eta tgt tcc acc ate gtt cag EL3 GMQN^TSL NF^S T I V Q ggg atg cag aac acg fcg eta aat tcc acc ate gtt cag N1QD DGDMDLSVS gat ggc gac atg gac ctg age gtc age D124N DG D S VS gat ggc gac atg aac ctg age gtc age

FIGURE 2.3

79 Figure 2.4 ASBT has only one A^glycosylation site occurring at Asn^°

(A) Af-glycosylation of ASBT. Transient transfection was performed as described

in the “Experimental Procedures”. The transfected COS-1 cells were incubated at

37 °C for 48 h in the presence (+) or absence (-) of /V-glycosylation inhibitor

tunicamycin (TNM). Lysates were subjected to SDS-PAGE, electroblotted onto

PVDF membrane, probed with rabbit anti-ASBT antibody and visualized by a chromogenic Westem Blotting kit (Vector Lab). (B). Cell surface expression of

ASBT wt and mutants. Detected by streptavidin conjugated alkaline phosphatase after cell surface biotinylation, immunoprécipitation, electrophoresis and blotting.

Arrows indicate the expected molecular weight of glycosylated (41 kDa)and nonglycosylated (38kDa) ASBT. Mock, pCMVP; WT, wild-type.

80 Mock wt N10D N328D N10D/N328D 1 I 1 TNM

B Mock wt N10D N328D N10D/N328D

*«ws>9W

FIGURE 2.4

81 Figure 2.5 Uptake activity of ASBTs from which the endogenous AAglycosyiation consensus sites have been removed

^H-Taurocholate uptake in COS-1 ceils transfected with pCMVS vector containing the indicated cDNA constructs. Uptake medium was spiked with 5 pM [^H]TCA

(0.2Ci/mmol). Results were given as percent of mutant uptake relative to wt

(81.3±5.7 pmol.Min'Vmg protein'^). Each bar is the m eantS.E. of three to four different measurements,wt, wild-type.

82 140 1

120 -

100 -

80 -

40 -

20 -

Mock Wt N10D N3280 N10/328D

FIGURE 2.5

83 Figure 2.6 Analysis of aglyco-ASBT with the engineered AAglycosylation sites

(A) Western blotting results on whole cell lysate; (B) cell surface expression of

ASBT wf and mutants. Arrows indicate that the AAglycosylation occurs on the inserted site.

84 Mock wt ELI EL3 D124N N10D 1 I 1 1 I 1 TNM + + + - + -

41 kD a-

B Mock wt ELI EL3 D124N N10D

41 kD a.

FIGURE 2.6

85 Figure 2.7 Uptake activity of agiyco-ASBTs (N10D) in which the AAglycosylation consensus sites have been inserted

^H-Taurocholate uptake in COS-1 cells transfected with pCMVS vector containing the indicated cDNA constructs. Uptake medium was spiked with 5 pM [^H]TCA

(0.2Ci/mmol). Results were given as percent of transport of the mutants relative to that of the wild-type (81.3±5.7 pmol.Min'\mg protein'^). Each bar is the meantS.E. of three to four different measurements,wt, wild-type. ELI, the engineered sites located in extracellular loopi between residue 113 and 116;

ELI, the engineered sites located in extracellular loop3 between residue 266 and

270.

86 120 1

100 -

80 - I 1 6 0 - g

40 -

20 -

Mock wt N10O N10D/D124N ELI EL3

Figure 2.7

87 CHAPTER 3

RESIDUES ESSENTAIL FOR SODIUM AND BILE ACID INTERACTION IN

ASBT

3.1. Introduction

The highly efficient enterohepatic cycling of bile acids is achieved by a vectorial transport system in the enterocytes as well as in the hepatocytes. This transport system has been identified composed of at least three distinct molecular components: active transporters responsible of the entry of bile acids to the cell, cytoplasmic bile acid binding proteins, and efflux carriers that pump bile acids out of the cell. Among them, the apical sodium-dependent bile acid transporter (ASBT) and the NaMaurocholate cotransporting polypeptide (NTCP) are the active transporters that uptake bile acids into enterocytes and hepatocytes [84,85}, respectively. These^twa transporters share a significant sequence similarity and belong to the same solute carrier subfamily (SLC10A).

ASBT has been cloned from hamster, rat, mouse, rabbit and human [85,86,88,

95] and found highly conserved among species (Figure 1.9).

To determine the structural basis for ASBT function, we previously demonstrated the presence of seven transmembrane (TM) spanning regions in

88 ASBT^by^W^ye@8ylali0n-SGann*nshmutagen6Si6'(Ghap*ef 2X AeGorëing-to-tlie- -

seven TM model, the extracellular regions of ASBT (Figure 3.1) Include the N-

termlnal, one small extracellular loop (EL2) and two long extracellular loops (ELI

and ELS), both of which, with their underneath transmembrane regions, are the

most conserved regions in the sub-family of bile acid transporters. It was

hypothesized that both regions are likely to be involved in the binding and translocation of sodium and bile acids [141]. Thus far several studies on ASBT

have been conducted to probe its functional regions essential for substrate binding and transiocation. Previous chemical modification of ASBT by using amino acid-specific reagents [142] and construction of a predictive 3D OSAR pharmacophore model [69] predicted several important interactions between bile acid and ASBT. Recently, the approach using side-specific reagents [139] suggested that Cys^^° may be in a sodium sensitive region; while Kramer’s group

[140] suggested that the cytoplasmic C-terminus with the connected transmembrane segment of ASBT may be the candidate regions for bile acid binding. However, no specific residue essential for transport function has been identified unequivocaiiy.

In this study, we substituted the charged and conserved residues in extracellular regions (according to the 7TM model) of ASBT with alanine. This study has led to the identification of the residue and region important for bile acid recognition and sodium interaction.

89 -3.2^ExpemmentakpfOcedufe8- ^ -

3.2.1 Materials

Sulfo-NHS-LC-BiotIn was purchased from Pierce (Rockford, IL). Antibody

against ASBT was obtained as described previously (Chapter 2). Antibody

against calnexin was obtained from StressGen (Victoria, BO Canada). [^H]

Taurocholate acid (50 Cl/mmol) was purchased from American Radiolabeled

Chemicals, Inc (St. Louis, MO). Unlabeled taurochollc acid and other bile acid

analogs were purchased from Sigma (St. Louis, MO).

3.2.2 Site-directed mutagenesis

Mutations were Introduced using a QuickChange kit from Stratagene (La

Jolla, CA) according to the manufacturer’s protocol. The pCMVS vector

containing human ASBT cDNA was a kind gift of Dr. Paul Dawson, Wake Forest

University (Winston-Salem, NO) and was used as the template In the PCR

mutation reactions. Mutagenesis primers (Table 3.1) were designed using the

Vector Suites software from InfoMax (Bethesda, MD) and synthesized by Operon

(Alameda, CA). Plasmids were purified by using a kit from Roche Molecular

Blochemlcals (Indianapolis, IN). All mutations were verified by sequencing In a

3700 DNA analyzer (Applied Blosystems) In the Plant-Mlcrobe Genomics Facility

at the Ohio State University.

90 COS-1 cells were obtained from the American Type Culture Collection and grown in Dulbecco’s modified Eagle's medium containing 10% fetal calf serum,

4.5g/L glucose, 100units/ml penicillin, and lOOpg/ml streptomycin (Life

Technologies, Inc.).

Transient transfection was conducted using the LipofectAMINE PLUS^" reagent (Invitrogen, Carlsbad, CA). Mock transfection with pCMVp was used as the control. On the day before transfection, trypsinized COS-1 cells were suspended in antibiotic-free medium and seeded in 24-well plates at a density of

6 x 1 0 ^ cell per well. On day 1 ,0 .4 pg/well of wild type or mutant plasmid was diluted by 21 pi of D-MEM, mixed with 4 pi of PLUS Reagent, and incubated at room temperature for 15 minutes; then the pre-mixed DNA-PLUS reagent solution (25 pi) was mixed with the 25 pi of the diluted LipofectAMINE^** reagent

(2.0 pi in 23 pi of D-MEM) and incubated at room temperature for 15 minutes; while the complex was forming, the culture medium on the cell was replaced by

200 pi of D-MEM containing 0.1 mM non-essential amino acids but no serum or arrtibtoticsr 50 pf of th r DNA-PLUS- LipofectAMINE^ reagent complex was added to each well, and incubated at 37 ° C at 5% CO 2; After 5 hours incubation, the medium was replaced with complete medium. On day 3 (48 hr after transfection), the transfected COS-1 cell were processed for the TCA uptake assay, western blot analysis and cell surface biotinylation.

91 —3;2;4-Btie-aetchiptake-asse

The transfected COS-1 cells In 24-well plates were Incubated In 0.2 ml of

uptake medium with the Indicated amount of [^H] TCA for 10 mln. After the

washing steps, the cells were lysed by 0.3 ml/well of PBS with 1% TrltonX-100,

then the aliquots were subjected to both liquid scintillation counting and protein

quantification by BCA protein assay kit (Pierce, Rocklord, IL). Uptake activity was

determined as the number of pmols of TCA taken up per mln and per mg of

protein.

For Inhibition study, the uptake of 5 [^H] taurochollc acid (S.P = 0.2 pci

/mmol) by transfected COS-1 cells was measured In the presence of Increasing

concentration of the respective bile acid analogues, namely 10'^ 10'^, 10*^, lOf,

10^,10 ^ or 10 ^ M. Uptake rates were plotted against Inhibitor concentration.

For sodium activation kinetics, the uptake of 5 pM [^H] taurochollc acid

(S.P = 0.2 pcI/mmol) was measured at seven extracellular concentrations of

sodium between 0 and 137 mM with Iso-osmolarlty maintained with choline

chloride.

The kinetic data were analyzed using nonlinear least-squares fitting by

SIgmaPlot software (Jandel Scientific). For bile acid uptake kinetics study, kinetic

constants were derived by fitting data with the MIchealls-Menten equation; V=

Vmax[S]/(Km+[S]). Where [S] Is the bile acid concentration, Vmax Is the uptake rate

at saturating [S], and Km Is the substrate concentration at 0.5 Vmax-

For uptake Inhibition study, the ICso for Inhibition was calculated by

competitive Inhibition kinetics: E=Emax(1 -[S]/([S]+IC 5o))- Where Emax Is the uptake

92 “ Off^HHawocholic-aeid-ln-the-ab9ence-©finhibHoiV“fSl-i»1he-lnhlbito^

concentration.

For sodium kinetic study, kinetic constants were derived by fitting data

with the Hiil equation: V=Vma%[Na*]"/(KNa" +[N aT)- Where [Na^ is the sodium

concentration, Vmax is the uptake rate at saturating [Na*], i.e. 137 mM, Kua is the

sodium concentration at 0.5 Vmax. and n is the Hill coefficient.

3.3 Results

in the extracellular side of ASBT, there is twelve highly conserved charged

residues (see the multiple sequence alignment of ASBTs from different species,

including rat, mouse, human. Figure 1.9), and their putative locations, according

to the 7TM model, are shown in Figure 3.1. Site directed mutagenesis on these

residues was conducted to evaluate their roles in sodium and substrate

interactions. Glu^^ was also selected since the negative charge is conserved in

this position. In fact, Asp’“ , Lys’“ , Arg“ ®, Glu“ ’ and Glu“ ^ are also conserved

in NTCP. After transient expression of the wild type and the alanine replacement

mutants in COS-1 ceils, protein expression and uptake of [®H] TCA were

monitored.

3.3.1 Immunoblots

Both the anti-ASBT and anti-calnexin were used in western blot study. As

an abundant ~90kDa resident endoplasmic reticulum membrane protein, calnexin

is believed to be expressed consistently under each experimental condition and

93 -therefere-ean-be-used-as-ft^positive-eentret-in-thls-studyrA&sliewn'iFKFigure

3.2A, the blot probed by anti-calnexin revealed single band around 90 kDa with

consistent intensity in all of the cell extracts including the mock-transfected cells,

while the blot probed by anti-ASBT showed two strong bands corresponding to

the molecular weights (41 and 38kDa, Figure 3.28) of the glycosylated and

unglycosylated forms of ASBT, evidenced previously (Chapter 2). The specificity

of anti-ASBT was verified by the absence of band in the mock-transfected ceils.

No K191A mutant could be detected in the transfected COS-1 cells, while

Calnexin from the same protein sample was easily detected. This result suggests

that Lys'^^ may be important for protein stability. Immunoblotting on the whole

cell lysate of the other mutants indicates that these mutants all expressed

normally.

In order to evaluate if the mutation influences the trafficking of the

transporter to the cell membrane, cell surface biotinylation of the transfected

COS-1 cells was conducted using a membrane impermeable biotinylated

reagent, Sulfo-NHS-LC-Biotin. As shown in Figure 3.2C, similar amounts of wild-

type and each of the tested mutant proteins were detected in cell extracts after

biotinylation. The specificity of the biotinylation reaction was verified by the fact

that no band could be detected when COS-1 cells transfected by wild-type ASBT

underwent all the labeling steps without the presence of Sulfo-NHS-LC-Biotin. To

make sure that Suifo-NHS-LC-Biotin is indeed a membrane impermeable reagent

in the system used here and thus only labels proteins on the cell surface, we

have conducted a parallel cell surface biotinylation experiment, followed by

94 -immun0pfeGipiWk)A^using^an*i-GalnexinrN0^bk)*inylate€k€alnexin^eoul€kbe------

detected (data not shown), whereas this abundant protein was easily detectable

in the whole ceil lysates.

Taken together, these results indicate that the all of the mutations except

K191A do not, or at least not significantly, alter the cell-surface expression and

trafficking of the transporter.

3.3.2 Uptake kinetics of ASBT mutants

Preliminary studies showed that [^H] TCA uptake in COS-1 cells

transfected by wild-type ASBT was linear for up to 15 min at 37 °C. The initial

screening on mutant uptake activity compared to wild-type was performed under

low [^H] TCA concentration (5 pM) with the presence and absence of sodium

(Figure 3.3). The strict sodium dependent TCA uptake was observed for all of the

functional mutants as well as wild-type, since only background uptake could be

detected for wild-type and all the mutants when sodium free uptake medium was

used (data not shown). Several loss-of-function mutants were identified, including

D120A, D122A, R256A, E261A (4.5% wt activity) and E282A (2.0 % wt activity).

Since immunoblotting study has revealed that all of these mutant proteins have similar protein expression level on cell surface as wild-type, the inactivity of these mutant must not be due to the defect in synthesis and trafficking, but due to the mutation-induced changes on the functional relevant regions of ASBT. K185A also shows a significant decrease in uptake activity (12.6% wt activity). Other mutants range from 39.9% (E281 A) to 141.1% (R254A) of the wide-type activity.

95 “ Tafurtherc haracterize the- potentfahrotes-gactr chargechrestctoes pteyairr

substrate interaction, we performed sodium-dependent TCA uptake kinetics

study on all of the functional mutants including E23A, D124A, K185A, K189A,

R254A and E281A. The kinetics constants are summarized in Table 3.2. The

apparent Kt value obtained for wild-type ASBT (11.3 pM) is consistent with the

values (17 or 13pM) reported previously [86,92]. However, the determined Vmax

value (312 pmol.min'\mg protein'^) in our study is much higher than the value

(96 pmol.min'Vmg protein'^) from COS-1 ceils transfected by DEAE-dextran

method [86], this is likely due to the higher transfection efficiency by

Lipofectamine method used in our study. E23A, K189A and R254A show Kt and

Vmax values comparable to those of wild-type, suggesting that they are not likely

involved in the substrate recognition/binding. Among the other functional

mutants, K185A, D124A and E281A, both K185 and E281A have slightly increased Kr values and have 16% and 42 % wiie-type Vmax values, respectively.

While D124A is the only mutant with both Kr and Vmax significantly different from wild-type. From this TCA uptake kinetics, we hypothesize that Asp^^^ may be in the bile acid recognition site.

To further test this hypothesis and determine the potential role Asp'^^ may play in bile acid and ASBT interaction, we determined the affinity of a group of other bile acid analogs to D124A. This was achieved by the inhibition study, in which [^H]-TCA uptake was compared between wild-type and mutant at increasing concentrations of each bile acid analog, including cholic acid (CA), deoxycholic acid (DCA), chenodeoxycholic acid (COCA) and

96 tauroelienodeoxyelioli&aGkt-fT€06A)-(stfueture9-shewFhii^ the Fi§ure3i4K

Besides wild-type and 0124A, we also incorporated K185A and E281A in the

study as the controls. Figure 3.5 is a representative inhibition curve by TCDCA,

in which D124A shows a clear right shift from wild-type, K185A and E281A

mutants. This shift reflects a more than 10 fold increase in ICSO, suggesting that

the binding affinity of TCDCA to the transporter drops evidently after the alanine

substitution on Asp'^^. As summarized in Table 3.2, for all of the four compounds tested, the ICSO values from both K18SA and E281A mutants are largely comparable to those from wild-type, suggesting that these bile acid analogs still retain the similar affinity to both K18SA and E281A as to wild-type. However, compared to wild-type, much larger ICSO values from D124A were observed for the three out of four inhibitors: CA, COCA and TCDCA. Taken together, these results indicate that Asp^^'* is likely located in the bile acid binding site since the binding affinity of most of the bile acid analogues drop with the alanine substitution in this position, while both Lys'^ and Glu^^\ compared to Asp^^^, are less likely involved in the direct interaction between bile acid and ASBT.

3.3.3 Sodium activation kinetics

ASBT cotransports sodium and bile acid with a 2:1 Na*: bile acid coupling stoichiometry [103]. The coupling mechanism for Na*to transport remains unclear. To study if the mutant transporters have altered sodium-binding activity, we first compared the [^H] TCA uptake of each functional mutant under saturating Na^ concentration (137 mM) to that under low and insaturating Na^

97 conceNfa*k)n.(42^mM^ This-melhod-ha&been^used-pfeviousiy to-scfeen^lhe

mutants for the potential Na^ Interaction region [143]. Figure 3.6 shows the

uptake ratios of each mutant. Single mutant R254A was identified with

significantly lower ratio of 0.2, suggesting that the transport function of this

mutant, compared to wild-type and other functional mutants, is significantly

defective in low Na* concentration. The TCA uptake with increasing

concentration of sodium was then determined to elucidate sodium activation

kinetics. Figure 3.7 shows the sodium-dependent uptake profiles for wild-type,

E281A and R254A. As can be seen, E281A gives a curve overlapped by that of

wild-type. E281A is shown here to represent all the other mutants except R254A,

all of which have the sodium-dependent kinetics undistinguishable from that of

wild-type. In contrast, R254A shows a clear shift in apparent sodium affinity.

Based on these data. Hill equation was used to estimate the kinetics parameters.

The results are shown in the inset \n the Figure 3.7. For the wt, the fitted n value

is equal to 2.1, which correlates well with the 2:1 stoichiometry of sodiumibiie acid for ASBT [103]. For R254A, the n value is 3.0, which reflects the change on the apparent sodium affinity of R254A. These data suggest that the region near

Arg^^ is a candidate region for sodium interaction.

3.4 Discussion

An important step in elucidating transport mechanism of transporter in the molecular level is to identify the specific amino acid residues involved in the substrate interaction. It is generally believed that the charged residues on the

98 extraGeUulafside'ef-the’transpefteFneFmaHy-may-not-be-essential^for-structurah

integrity, but are likely involved in initial substrate recognition or binding. With the

understanding of the membrane topology of ASBT (Chapter 2), which predicts

the potential 2-D topological location of each residue, a rational design of an

extensive mutagenesis experiment becomes feasible. In this study, we focused

on the conserved charged residues on the extracellular side of ASBT by alanine-

scanning mutagenesis, since these residues, at least some of them, may be

functionally relevant and thus are appropriate candidate sites to start with.

From the functional analysis of the library of 12 Ala-substituted mutants,

we were able to discern whether a residue is involved in bile acid recognition or

not.

Among the mutation sites leading to functional mutants, Glu^, Lys^^® and

Arg^^ are not likely located in the bile acid recognition site, since mutations on these sites do not change TCA uptake of ASBT, as evidenced by comparable kinetic constants after each mutation (Table 3.2). Identification of a group of loss- of-function mutants indicates that residue Asp’®’, Asp’“ , Arg^®®, Giu“ ’ and

Glu^®^ are likely the functionally important sites in the transporter. Along with three previously identified loss-of-function polymorphisms in human ASBT genes

[86,89], including L243P, T262M and P290S, all of these mutation sites are in either ELI or EL3 or the connected transmembrane domains. These results further substantiate the importance of these regions for transport function [141].

However, it remains unclear whether these residues localize in bile acid recognition or binding sites or not. This is because at least two reasons can

99 -accoun^fühthe-loss^-functton-oFthe-transpofterrheT-l-y^thfrsobstittition-ofarr

essential residue in the binding site and 2) the mutation induced long-range

conformational change in the binding site. Therefore, generation of the functional

mutants on these sites might be necessary to further define the exact functions of

these residues.

A single functional mutant, D124A, was identified with TCA uptake kinetics

significantly different from wt (Table 3.2). From the inhibition study, D124A

mutant also showed lower affinity to several other bile acids (Table 3.3). Taken

together, these data suggest that Arg^^^ (on EL1) is likeiy involved in the bile acid

recognition. Furthermore, we hypothesize that there is a hydrogen-bond

formation between Asp^^^ and 7a OH in bile acids: The carbonyl group on Asp is

the H-bond receptor and the 7 OH group is the H-bond donor. This

hypothesis is supported by the following evidences: (1) Previous pharmacophore

model [69] predicted that either 7 or 12 OH provide a H-bond donor feature

essential for the affinity; (2) From our uptake kinetics and inhibition study, when

Asp^^* is substituted by Ala, the ICSO values of bile acids containing 7-OH,

including taurochollc acid, cholic acid, chenodeoxycholic acid and

taurochenodeoxycholic acid, increase dramatically (Table 3.2 and 3.3),

suggesting lower affinity between bile acid and ASBT because of the absence of the hydrogen bond after the mutation; (3)in addition, for the bile acid not containing 7-OH group (deoxycholic acid), ICSO values and binding affinity

remain insensitive to the removal of the carbonyl group since such H-bond does not exist in the first place. To test this hypothesis, mutant D124N that retains the

100 hydrogen=forming'carbonyt‘groaFrwas‘constructed~and'subjecte(ftaTCA^uptaker

kinetic study. However, D124N still has a lower apparent binding affinity for

taurochollc acid but a Vmax comparable to that of wild-type (Table 3.2).

Nevertheless, we do not think that this result necessarily disapproves our

hypothesis, since the substitution of Asp by Asn may disturb local conformation

in the substrate recognition site that can affect the H-bond formation as well.

According to the pharmacophore model for ASBT [69] and previous experimental results [115], an e-amino group of a lysine residue was predicted as a hydrogen donor to interact with the negatively charged bile acid side chain.

Based on the 7TM model for ASBT, there are three extracellular Lys residues on position 185,189 and 191. Among them, Lys'^^ seems unlikely involved in substrate interaction as reasoned above; Lys'^^ might be important for maintaining the correct conformation for ASBT, since no K191A is expressed in the transfected COS-1 cells; however, the substitution of Lys^^ to Ala changed the uptake kinetics with a slightly decreased apparent TCA affinity, and significantly reduced apparent uptake capacity (Table 3.2). Taken together, we anticipate that the e-amino group of a Lys^^ might be the previously proposed hydrogen-bond donor. As shown in Figure 3.8, the putative H-bond formations between ASBT and bile acid are illustrated. These interactions may contribute partially to the recognition affinity between transporter and its substrate. It is also interesting to note that this mapping of the hydrogen-bond acceptor and donor features to Asp^^^ and Lys'^ respectively appears biologically sound since these two residues are adjacently localized. In addition, this designation of the

101 - prectetect-ehernieaMeatures-to-speeifie-aminchaeid-resieiues^in-AS&T

demonstrates the biological relevance of the previous pharmacophore model.

As for the sodium interaction site, we were surprised to see that the

mutant with Ala substitution on the positively charged Arg^^ was identified with

the altered apparent sodium affinity while retaining full wild-type activity at

saturating sodium concentration. It is generally believed that for sodium-

dependent active transporter, sodium binding triggers the conformational change essential for substrate recognition and translocation. Therefore, we hypothesize that Arg^^ is located in sodium sensitive region that is important for either sodium binding or the coupling of sodium to bile acid transport. When sodium concentration is high and saturating, the mutation-induced defect in the ability to bind sodium or to couple sodium to bile acid binding might be negligible and thus hidden, whereas this defect becomes evident while sodium concentration is low and insaturating. The identification of sodium interaction site to the region around

Arg^^(EL3) is in a line with the previous finding, in which Cys^^°, another residue in EL3, was also suggested to be in a sodium sensitive region [139].

The localization of the bile acid recognition site to the Asp'^^ (ELI) and

Lys'^(EL2) seems contradictory with the recent finding [140], which favored the regions including TM7 and C-terminus as the potential bile acid binding sites. Our explanation to this apparent inconsistency is that ASBT has several sites for bile acid recognition or binding, in which bile acid resides sequentially in a dynamic recognition and translocation process. Depending on the rationale of the experimental approaches, different methods may have varied resolution and thus

102 -generate-apparefrtly-distinet-FesuHshBiie-aeid-may-tnttially'interact^witMh»

extracellular residues such as Asp'^^ and Lys'^ (Identified In this study), and

eventually Interact with the TM7 and the C-termlnal of ASBT (Identified recently

[140]) during an unknown translocation process.

In summary, based on the 7TM membrane topology model, Ala-scanning

mutagenesis allows us to Identify the potential Interaction sites for bile acid and

sodium. Future studies with the elucidation of helix packing should lead to a

further clarification of the transport mechanism.

103 E23A CA TOG TGT GTG GTA GOT GOG AGG AAT TTG AAT AAC D120A GG TAT TGG GTA GGTGGG GAG ATG GAG GTG AG D122A GG GTG GAT GGG GOG ATG GAG GTG AG D124A G GAT GGG GAG ATG GCA GTG AGG GTG AGG ATG K185A GGA ATG TTT GTT AAT GAG GCG TGG GGG GAA AAA GCA AAG K189A GAG AAA TGG GGG GAA GCA GGA AAG ATG ATA GTT AAA ATT GG K191A GG GGG GAA AAA GGA GCG ATG ATA GTT AAA ATT GG R254A GT GTA GGG TGG TAG GCG TGG GGA AGG GTT G R256A G TGG TAG AGG TGG GCA AGG GTT GOT TTT GAA AGG G E261A G GGA AGG GTT GGT TTC GCG AGG GGG ATG GAG AAC E281A GTG TGG TTC ACT CGT GCA GAA CTG AAT GTG GTA TTG AG E282A G TGG TTG ACT GGT GAG GCC GTG AAT GTG GTA TTG AG

Table 3.1 Sequences of mutagenesis primers.

104 Extracellular

NT

ELI EL3

EL2 till w

TM2 TM3 TM4 TM5 TM6 TM7TM1 7TM

80 ISO 160 218 227 310

IL2 IL3 CT IL1

Figure 3.1 Putative seven-TMs topology model for ASBT

The naturally occurring AAglycosylation site is indicated by a Y-shaped symbol

Residues subjected to Ala-scanning are marked in open ellipsoids. EL, extracellular loop; IL, intracellular loop; NT, amino-terminal; CT, carboxyl- terminal.

105 Mock wt 23 120 122 124 185 189 191 254 256 261 281 282 kDa

z a C...Ï* mm s-r ■ ' ** - 90

B W# A #m am m m ^ É É

Mock wL 23 123.122 124 131135 13S 234 233 2 3 1 2 3 1 232

^ "** W”* «T-ja» w im mii, I ost fc,-v. -41 -3 8

Figure 3.2 Immunodetection of ASBT wild-type and Ala-substituted mutants

After SDS-PAGE and blotting, the PVDF membrane was out into two pieces along the line indicated by the 71KD molecular weight maker. The high and low mw parts were probed by anti-calnexin (A) and anti-ASBT (B) antibodies, respectively; (C) Cell surface expression of wf and mutants.

106 Jrnax(pmol.min\ Kt (p M) mg protein ’) Wild-type 11.3±1.9 312.5± 12.8 E23A 11.6 ±2.0 400.4 ±17.8 D124A 48.5 ± 6.2 1004.4 ±50.9 D124N 30.1 ± 6 .0 361.8 ±24.2 K185A 13.4 ± 1 .8 52.9 ± 1.8 K189A 1 1 .3 ± 1 .4 430.4 ±12.9 R254A 1 1 .5 ± 1 .6 391.6 ±14.2 E281A 19.6 ± 4 .5 131.6 ± 9 .0 E282D 12.3±3.4 104.4 ±7.5

Table 3.2 TCA uptake kinetic constants of wild-type and mutants

COS-1 cells were transfected with the wild type and indicated mutant ASBT transporters as described under “ Experimental Procedures.” Uptake was measured at TCA (0.02 Ci/mmol) concentration ranging from 0 to 150 pM

(measured in triplicates). The kinetic data were analyzed using SigmaPlot software (Jandel Scientific). Kinetic constants were determined by nonlinear least-square fitting procedures and reported as the fitted values.

107 ü) b î a [3H] TCA uptake ( % wt ) I [I ê S 8 8 8 8 8 g - 1 ■ I - I - I- î Mock wt 1 I S E23A t I D120A T3 S’ I D122A D124A s I I K185A I K189A K191A i I g R254A

« ! R256A 8 E261A I (D g E281A m m m m

(D I E282A 1+ I 10 H

Bile acids R I (7) R2(12) R3 TCAOH OH NHCCHzhSOsH CA OH OH OH DCA HOHOH COCA OH H OH TCDCA OH H NH(CH2)2S0 3 H

TCA: taurocholic acid CA; cholic acid DCA: deoxy-cholic acid COCA: chenodeoxy-colic acid TCDCA: taurochenodeoxy-choiic acid

Figure 3.4 Structures of bile acid analogues

109 :

120

S 100

WT D124A K185A E281A

10- 10- 1 0 " 10* * 10^ 0.0001 0.001 0.01 [Taurochenodeoxycholic acid] M

Figure 3.5 Inhibition of TCDCA on H-TCA uptake of wiid-type and mutants

^H-TCA uptake was measured for wt or mutant at the increasing concentration of each inhibitor at the fixed concentration of ^H-TCA (5pM). The curves were fitted by competitive inhibition type kinetics: E=Emax(1 -C/fC+ICgo)), performed in

SigmaPlot using nonlinear least-squares fitting.

110 IC 5 0 (mM)

CADCACOCATCDCA (7,12-OH) (12-OH) (7-OH) (7-OH)

WT 58±9 34±10 19±4 1 1 ± 2 D124A 128±27* 44±11 70±6* 154±33* K185A 71±18 58±18 27±6 15±3 E281A 75±15 48±14 16±4 15±5

Table 3.3 Summary of IC 50 values of each bile acid analogue to wt and mutants

*: p < 0.05. student t-test.

I l l 0.7

0.6 - ►I4 I E 0.5-

♦ m 0.4 E CM 0.3- SrSSS*

^ 0 2 -

0.1 - M 0-^ wt E23A D124A K18SA K189A R2S4A E281A E282D

Figure 3.6 Sodium dependency of [^H]-TCA uptake by wt and mutants

Uptake of [^H]-TCA into the transfected COS-1 cells was measured at sodium concentration of 137 mM and 12 mM. The results are given as the ratio between the uptake activity in low (12mM) and high (137mM) sodium concentration.

112 Figure 3.7 Sodium activation kinetics of wt and mutants

^H-TCA uptake into transfected COS-1 cells was measured at increasing Na+ concentrations using choline chloride as an equimolar replacement for NaCI. The curves were fitted by Hill equation: V = V^axINa]" /([Na]%KNa"). Every data point represents the means ± S.E of three to six measurements.

113 120

100

R254A 80 i E281A 60

40

20

0 0 20 40 60 80 100 120 140

Inset KN.(mM) n

Wild-type 10.4±0.8- ^ 2,1 -± 0.3

E281A 12.3 ± 1 .7 2.2 ± 0 .5

R254A 22.4 ±1.3 3.0 ± 0.5

Figure 3.7

114 2 4 ^ 0 -

H \ NH \ Lys'.185 I ASBT 124

Figure 3.8 Putative interactions between ASBT and bile acid

The carbonyl group in serves as a H-bond receptor to form a H-bond with the 7a-0H group in bile acid; the e-amino group (in Lys^^) acts as a H-bond donor to form H-bond with the negatively charged side chain in bile acid molecule.

115 CHAPTER 4

FUTURE DIRECTIONS

This dissertation provides strong experimental evidence supporting a 7-

TM model for ASBT and reveals a group of extracellular charged residues

essential for substrate recognition and transport function. This work will serve as

a starting point for rational design of experiments to further study the transport

mechanism of ASBT. We expect that future work In the following general areas

will further our knowledge of this transport system.

4.1 Helix packing of transmembrane segments of ASBT

N-glycosylatlon scanning and site-directed mutagenesis has allowed elucidation of the membrane topology of ASBT (Chapter 2) and delineation of amino add residues essentlaf for substrate^ recognition or transporter function

(Chapter 3). However, high-resolutlon structural and dynamic Information Is required to understand the transport mechanism and substrate translocatlon pathway. In the absence of a crystal structure, determination of the tertiary contacts and proximity between transmembrane domains of ASBT, I.e. helix packing. Is a logical extension of this dissertation work.

116 " "The-techniques asechto-stody-hetfarpacking (rf~membrang proteins arg ------

largely derived from the pioneering work on the lactose permease ofEscherichia

coli by Kaback’s group. The approaches Include disulfide crosslinking [144], site-

directed chemical cleavage [145], site-directed excimer fluorescence [52] and

site-directed spin labeling [53]. For ASBT, the task is to extend these approaches

to show in situ how the seven transmembrane segments align and juxtapose in the lipid bilayer.

4.2 Further verification of biological relevance of pharmacophore models

Baringhaus' group reported a pharmacophore model for ASBT inhibitors and substrates [69]. In their study, a training set of dozens of chemically diverse

ASBT inhibitors were used to derive common chemical features essential for transporter affinity. According to this pharmacophore model, the substrate or inhibitor with the strongest affinity should have one hydrogen bond donor, one hydrogen bond acceptor and three hydrophobic features (Figure 4.1 A). This model matches very well with the structure of natural ligands (bile acid) for ASBT

(Figure 4.1 B). To be specific, bile acid compounds have four out of five features predicted by this model: a) one of the alpha-oriented hydroxy groups at either position 7 or 12 fits the H-bond donor feature; b) the negatively charged site chains serve as the hydrogen bond acceptor; c) the 0 ring and methyP^ map the one hydrophobic feature and methyl^^ maps the second hydrophobic feature.

Theoretically, this pharmacophore model is complimentary to the substrate recognition site of ASBT. in fact, our experimental results (Figure 3.8)

117 suggested-that-Asp^and-tys^in-ASBT-may-matelrthe-predicted-K-boneh

receptor and donor features, respectively. It becomes obvious that further studies

are necessary to identify the other two potential hydrophobic recognition sites in

ASBT. It is anticipated that several hydrophobic amino acid residues in either the

extracellular loops or the transmembrane domains or both are involved in these two interaction sites. Since there are a large number of conserved hydrophobic

residues in these regions of ASBT, it does not seem reasonable to mutate every of them to probe the residues essential for the hydrophobic interaction. However, the docking results from a predictive 3-D model (Chapter 4.3) and the structural information of helix packing (Chapter 4.1) will greatly facilitate and rationalize the selection of the mutation sites.

4.3 Construction of a comprehensive structural and predictive model of ASBT

As discussed in Chapter 1, molecular modeling is rapidly becoming a feasible alternative to obtain structural information of membrane transporters in the absence of crystal structures, especially due to the rapid evolvement in the fields of computational biology and bioinformatics.

Since no X-ray structures on ASBT or other SLC proteins have been reported so far, construction of a 3-D model of ASBT will be based on homology modeling in combination with other essential protein modeling techniques. To search for sequences with high sequence homology to (parts of) ASBT, a BLAST search has been performed against the Protein Data Bank (PDB) at RCSB f htto://www. rcsb.oro/Ddbl. As expected, no homology was detected between

118 ASBT-and-pfOtelns-wfth-high-fesolotion-stfticttifes-stored-ln-PDBv The-onljr crystallized Integral membrane protein with a 2D membrane topology similar to

ASBT Is bacterlorhodopsin (BR), a photon-driven hydrogen Ion pump. Both

ASBT and BR have a trans location of N and C terminus (Nex^Cmt) and seven transmembrane(TM) segments. Therefore, the TM-domalns of ASBT could be modeled, using BR as a template. It should be noted that the biological relevance of the model of TM regions of ASBT should be verified by experimental approaches, such as which have been used to determine the Inter-dlstance and packing of the TM domains. Based upon a verified model of the TM regions, the connecting loops and terminal domains can be further modeled using other template structures and structure prediction techniques/protocols.

A complete 3D-model of ASBT thus constructed can be used as a working hypothesis to direct further studies. Including:

(1) to determine the substrate-binding site using "docking" methods, which probe the molecular surface of ASBT with a natural substrate. I.e. cholic acid or taurocholic acid, to select a potential binding site on the basis of the lowest complex energy.

(2) to compare the predicted binding site with the sites revealed by experimental techniques. The biological relevance of the model Is thus assessed, and the model may need to be modified.

(3) to create point mutations at sites putatlvely essential for transport function by experimental approaches, and simulate the effects of these mutations on the overall three-dimensional structure.

119 (4)-to-8ifTttrtatfrconfofmattenahchange8 triggered bysodttirr^bincKngr

reveal the coupling mechanism of sodium and bile acid transport.

4.4 Crystaliization of ASBT or structurally related transporters

Among all of the strategies for structural elucidation of proteins, obtaining

a crystal structure for the protein is clearly the most attractive and effective

choice. However, for most integral membrane proteins, this approach is also the

most cumbersome one, due to the inherent difficulties with the expression,

purification and crystaliization of membrane proteins

Presently, the only way to obtain high resolution data is by X-ray or

electron diffraction, and only a handful of membrane protein structure have been

resolved to < 3 A. For membrane proteins, the rate-limiting step is finding

crystallization conditions for the detergent-solubilized protein. Although still

extremely difficult, some recent successes afford considerable promise. The surface of bacterlorhodopsin was revealed at 3.0 A resolution by high-resolution electron crystallography [146]; the molecular basis of channel selectivity of the glycerol channel (GIpF) was elucidate by a 2.2 A structure [147]; and recently, the structure of a homologue of the multidrug resistant ATP binding cassette transporters was determined by x-ray crystallography to a resolution of 4.5 A

[148]. In this study, approximately 96,000 crystaliization conditions using about

20 detergents were tested to yield crystals with good quality for x-ray structure determination.

120 ——Wi*h-*he^c0nlinu0U8-pfgg#e666&in-x-Fay^mys*aWGgfaphy, eleelfon^^ ------

crystallography and nuclear magnetic resonance (NMR) spectroscopy, and

implementation of novel technologies, we expect that more high-resolution

structures of membrane proteins (including ASBT) will be resolved at an

accelerating pace in the near future.

4.5 Membrane topology and structure-function relationship of NTCP

As stated previously, both ASBT and NTCP are strictly sodium-dependent,

and believed to share the similar membrane topology and 3D structure due to

their high sequence homology (Figure 1.7,1.8 and Figure 4.1). Therefore, the

results we obtained for ASBT in this dissertation work can be, at least partially,

extrapolated to NTCP. To be specific, NTCP is likely to have seven

transmembrane segments as ASBT, and the functional regions for NTCP would

also situate in locations similar to those of ASBT. However, these hypotheses

need to be verified, and can be achieved by similar methods applied to ASBT. It

is worth noting that studying ASBT and NTCP in parallel may provide us a

possible way to validate the predicted structure- activity relationship of these two

transporters. For example, if sodium and bile acid follow a distinct translocation

pathways, we can substitute the putative sodium interaction/translocation region

in ASBT with the corresponding region in NTCP to create a chimeric transporter,

which, if is still functional, will provide strong experimental evidence to support

that the putative sodium binding/translocation regions is correct and are

interchangeable between ASBT and NTCP; similarly, since early studies have

121 -9howfMhat-NTGP^h@s^»moch-bfoaderstibstrate spectficity thaiTASBT[90921; t h r

substitution of the substrate binding/translocation of ASBT with that of NTCP may

lead to a sodium-dependent mutant ASBT with a broader substrate specificity.

4.6 Potential protein interaction between ASBT and ILBP

Besides the presence of active transporters, such as ASBT and NTCP, in

the enterohepatic circulation, the soluble binding proteins in the cytoplasm of

hepatocytes and ileocytes are also very important for efficient cycling and

transport of bile acids. The ileal lipid-binding protein (ILBP), a 14-kDa protein,

has originally been identified as a major bile acid binder in the cytosol of ileocytes

(Figure 1.5) [149-151]. Photoaffinity labeling experiments suggested that ILBP is

the only physiologically relevant bile acid-binding protein in the ileal cytosol and

specifically interacts with the ileum ASBT at the cytoplasmic face of the ileocyte

[152-154]. Recently, it was reported that the incorporation of photolabile bile

acids into ILBP is stimulated by the presence of bile acids instead of being

inhibited [155]. This suggested that there is a substrate load modification

mechanism for the binding of bile acids to ILBP. More recently, the bile acid-

binding sites of the ILBP was mapped and identified [156].

Radiation-inactivation analysis revealed that the size of the functional

transporting unit for Na*-dependent taurocholate uptake was 451 +/- 35 kOa,

furthermore, proteins of 93 kDa (the ASBT dimers) and 14 kDa (ILBP) were

identified as putative protein components of this bile acid active transport system

[154]. Therefore, it is reasonable to hypothesize that ASBT and ILBP function

122 -together-as-part-of-a-comptex-in-the-Heeeytesr-We-beKeve-the-Hltistratiofhof the-

potential Interaction between ASBT and ILBP is one of the important aspects in

the understanding of the intestinal bile acid active transport.

There are numerous techniques available to study protein-protein

interactions, such as the conventional two-hybrid methods and its descendants,

fluorescence resonance energy transfer methods, protein mass spectrometry,

evanescent wave methods, molecular modeling and so on [157-164]. Recently, a

reverse Ras recruitment system (reverse RRS) approach was specifically

designed for the identification of protein-protein interaction with integral

membrane proteins, such as receptors, ion channels and transporters [165]. In

this approach, the membrane protein of interest is used as bait and expressed in

its natural environment, the membrane, whereas the protein partner (the prey) is

fused to a cytoplasmic Ras mutant. Protein interactions between the bait and

prey proteins leads to Ras membrane translocation and activation of viability

pathway in yeast.

If this reverse RRS approach is applicable to the case of ASBT and ILBP,

we will be able to test our hypothesis. If we verify that these two proteins are

subject to transient interactions, we will incorporate site-directed mutagenesis to

identify the potential interaction sites on both proteins. If some mutations do

effect protein interactions, we will seek for an appropriate system, such as

contransfection of the two proteins in the polarized MDCK cell line, and study if

the mutations have clear measurable biological effects, i.e. the changed transport

rate.

123 >4.7 Final-Femarks--

It is clear that over the past decade we have gained a great deai of

knowledge about the function of membrane transporters and how they can be manipulated as drug targets for maximizing absorption and bioactivity. Although we are still without crystal structures for SLC proteins there has been an explosion in interest in structure-activity relationships. This has been seen largely as part of a wider growth in using computationai approaches for understanding in vitro data and in particuiar in the fields of drug design, optimization and AOME properties [2]. The resolution of the three-dimensional structures of solute transporter protein modeis is presumably low, however, they can be justified by their ability to confirm biologically relevant phenomena. As with all models, the continued input of novel experimental data and revalidation of the model may eventually lead to highly predictive systems capable of in silica detection and design of novel solute transporter substrates. It is likely that the integration of bioinformatics, computational and experimental techniques will drive our understanding of SLC protein to new levels and afford an opportunity for further possibie therapeutic targets and drug design opportunities.

124 HB Donor

'Hydrophobic

Hydrophobic

B

Figure 4.1 Pharmacophore model for ASBT(A) and structure for natural ligand (B)

125 Figure 4.2 Putative 7TAA model of Human ASBT and NTCP

Black circle represents conserved amino acid residues in all sodium-dependent bile acid transporters including ASBTs and NTCPs. The Y shape marks the N- glycosylation site.

126 BIBLIOGRAPHY

[1] M.H. Saler, Jr., A functional-phylogenetic classification system for transmembrane solute transporters, Microbiol Mol Biol Rev 64 (2000) 354-411.

[2] S. Ekins, C.L. Waller, P.W. Swaan, G. Cruciani, S.A. Wrighton, and J.H. Wikel, Progress in predicting human ADME parameters in silico, J Pharmacol Toxicol Methods 44 (2000) 251-272.

[3] R. Kerb, S. Hoffmeyer, and U. Brinkmann, ABC drug transporters; hereditary polymorphisms and pharmacological impact in MDR1, MRP1 and MRP2, Pharmacogenomics 2 (2001) 51-64.

[4] W. Sadee, V. Drubbisch, and G.L. Amidon, Biology of membrane transport proteins, Pharm Res 12 (1995) 1823-1837.

[5] M. Dean, A. Rzhetsky, and R. Allikmets, The human ATP-binding cassette (ABC) transporter superfamily. Genome Res 11 (2001) 1156-1166.

[6] J. Kyte, and R.F. Doolittle, A simple method for displaying the hydropathic character of a protein, J Mbf Biol 157 (1982) 105-132.

[7] M.G. Claros, and G. von Heijne, TopPred II; an improved software for membrane protein structure predictions, Comput AppI Biosci 10 (1994) 685-686.

[8] 8. Rost, P. Fariselli, and R. Casadio, Topology prediction for helical transmembrane proteins at 86% accuracy. Protein Sci 5 (1996) 1704-1718.

127 _____^^@_E^ygnadyrand-k-SimomrPnn€iple9-govefning'amimo^aGi(H ~ composition of integral membrane proteins: application to topology prediction, J Mol Biol 283 (1998) 489-506.

[10] J. Meuller, and J. Rydstrom, The membrane topology of proton- pumping Escherichia coli transhydrogenase determined by cysteine labeling, J Biol Chem 274 (1999) 19072-19080.

[11] J.C. Long, S. Wang, and S B. Vik, Membrane topology of subunit a of the FIFO ATP synthase as determined by labeling of unique cysteine residues, J Biol Chem 273 (1998) 16235-16240.

[12] T.W . Loo, and D M. Clarke, Membrane topology of a cysteine-less mutant of human P-glycoprotein, J Biol Chem 270 (1995) 843-848.

[13] M. Hollmann, C. Maron, and S. Heinemann, N-glycosylation site tagging suggests a three transmembrane domain topology for the glutamate receptor GluRI, Neuron 13 (1994) 1331-1343.

[14] Z.G. Wo, and R.E. Oswald, Transmembrane topology of two kainate receptor subunits revealed by N- glycosylation, Proc Natl Acad Sci U 8 A 91 (1994) 7154-7158.

[15] T. Michikawa, H. Hamanaka, H. Otsu, A. Yamamoto, A. Miyawaki, T. Furuichi, Y. Tashiro, and K. Mikoshiba, Transmembrane topology and sites of N- glycosylaticrv of inositof 1,4,5- trisphosphat» receptor, J BiolCherrv 269 (1994) 9184-9189.

[16] R.A. Schwalbe, Z. Wang, L. Blanchi, and A.M. Brown, Novel sites of N-glycosylation in R0MK1 reveal the putative pore- forming segment H5 as extracellular, J Biol Chem 271 (1996) 24201-24206.

[17] M. Popov, L.Y. Tam, J. Li, and R.A. Reithmeier, Mapping the ends of transmembrane segments in a polytopic membrane protein. Scanning N-

128 -glycosylation-mutagenosIs-ô^^extraGytoselie-leopfr-iFKtiie-anjon-exehangeF, ban& 3. J Biol Chem 272 (1997) 18325-18332.

[18] H.P. Adamo, A.J. Caride, and J.T. Penniston, Epitope mapping by deletion mutants reveals the transmembrane topology of the plasma membrane Ca2+ pump, Ann N Y Acad Sci 671 (1992) 415-417.

[19] K.M. Covitz, G.L. Amidon, and W. Sadee, Membrane topology of the human dipeptide transporter, hPEPTI, determined by epitope insertions. Biochemistry 37 (1998) 15214-15221.

[20] 0 . Kast, V. Canfield, R. Levenson, and P. Gros, Membrane topology of P-glycoprotein as determined by epitope insertion: transmembrane organization of the N-terminal domain of mdr3. Biochemistry 34 (1995) 4402- 4411.

[21] R. Koebnik, and V. Braun, Insertion derivatives containing segments of up to 16 amino acids identify surface- and periplasm-exposed regions of the FhuA outer membrane receptor of Escherichia coli K-12, J Bacteriol 175 (1993) 826-839.

[22] M. Sahin-Toth, R.L. Dunten, and H.R. Kaback, Design of a membrane protein for site-specific proteolysis: properties of engineered factor Xa protease sites in the lactose permease of Escherichia coli. Biochemistry 34 (1995) 1107- 1112.

[23] 0 . Manoil, Analysis of membrane protein topology using alkaline phosphatase and beta-galactosidase gene fusions. Methods Cell Biol 34 (1991) 61-75.

[24] J.P. Sarsero, and A.J. Pittard, Membrane topology analysis of Escherichia coli K-12 Mtr permease by alkaline phosphatase and beta- galactosidase fusions, J Bacteriol 177 (1995) 297-306.

129 -t25h\^Ar.“PrirvzrafKJ-JrBeekwtthrGene-fu8iorY-analy8irofrnembfane^ protein topology: a direct comparison of alkaline phosphatase and beta- lactamase fusions, J Bacteriol 176 (1994) 6410-6413.

[26] D. Jording, and A. Puhler, The membrane topology of the Rhizobium meliloti C4-dicarboxylate permease (DctA) as derived from protein fusions with Escherichia coli K12 alkaline phosphatase (PhoA) and beta-galactosidase (LacZ), Mol Gen Genet 241 (1993) 106-114.

[27] D. Boyd. B. Traxler, and J. Beckwith, Analysis of the topology of a membrane protein by using a minimum number of alkaline phosphatase fusions, J Bacteriol 175 (1993) 553-556.

[28] M. Sahin-Toth, A. Karlin, and H.R. Kaback, Unraveling the mechanism of the lactose permease of Escherichia coli, Proc Natl Acad Sci U S A 97 (2000) 10729-10732.

[29] Q. Wang, and H.R. Kaback, Location of helix III in the lactose permease of Escherichia coli as determined by site-directed thiol cross-linking. Biochemistry 38 (1999) 16777-16782.

[30] I. Kwaw, K.C. Zen, Y. Hu, and H.R. Kaback, Site-directed sulfhydryl labeling of the lactose permease of Escherichia coli: helices IV and V that contain the major determinants for substrate binding. Biochemistry 40 (2001) 10491- 10499.

[31] J. Wu, D. Hardy, and H.R. Kaback, Site-directed chemical cross- linking demonstrates that helix IV is close to helices VII and XI in the lactose permease. Biochemistry 38 (1999) 1715-1720.

[32] H.R. Kaback, Structure/function studies on the lactose permease of Escherichia coli, Acta Physiol Scand SuppI 643 (1998) 21-33.

130 " — "pSJ-PrBraonrBrPefSSorh-HrRrKatmckrand-Grvon-Hetjner Alaninfr------insertion scanning mutagenesis of lactose permease transmembrane helices, J Biol Chem 272 (1997) 29566-29571.

[34] J. Wu, and H.R. Kaback, Helix proximity and ligand-induced conformational changes in the lactose permease of Escherichia coli determined by site-directed chemical crosslinking, J Mol Biol 270 (1997) 285-293.

[35] S. Frillingos, and H.R. Kaback, Probing the conformation of the lactose permease of Escherichia coli by in situ site-directed sulfhydryl modification. Biochemistry 35 (1996) 3950-3956.

[36] J. Wu, S. Frillingos, J. Voss, and H.R. Kaback, Ligand-induced conformational changes in the lactose permease of Escherichia coli: evidence for two binding sites. Protein Sci 3 (1994) 2294-2301.

[37] M. Sahin-Toth, R.L. Dunten, and H.R. Kaback, The lactose permease of Escherichia coli: a paradigm for membrane transport proteins, Soc Gen Physiol S er48 (1993) 1-9.

[38] S. Frillingos, M. Sahin-Toth, J. Wu, and H.R. Kaback, Cys-scanning mutagenesis: a novel approach to structure function relationships in polytopic membrane proteins, Faseb J 12 (1998) 1281-1299.

[39] S. Frillingos, A. Gonzalez, and H.R. Kaback, Cysteine-scanning mutagenesie of helix IV and the^adjdnlng^loops in^ the lactose permease of Escherichia coli: Glu126 and Arg144 are essential, off. Biochemistry 36 (1997) 14284-14290.

[40] T.W. Loo, and D M. Clarke, Identification of residues in the drug- binding domain of human P- glycoprotein. Analysis of transmembrane segment 11 by cysteine-scanning mutagenesis and inhibition by dibromobimane, J Biol Chem 274 (1999) 35388-35392.

131 f4tJ-D:erReevesrEîNrQorenrMrHrAkabasrand- S:C: tummisv Structural and electrostatic properties of the 5-HT{sub3} receptor pore revealed by substituted cysteine accessibility mutagenesis, J Biol Chem 13 (2001) 13.

[42] E.S. Dawson, and J.N. Wells, Determination of amino acid residues that are accessible from the ligand binding crevice In the seventh transmembrane-spannIng region of the human A(1) adenosine receptor. Mol Pharmacol 59 (2001) 1187-1195.

[43] W. Xu, J. LI, 0 . Chen, P. Huang, H. Weinstein, J.A. Javltch, L. Shi, J.K. de Riel, and L.Y. LIu-Chen, Comparison of the amino acid residues In the sixth transmembrane domains accessible In the blndlng-slte crevices of mu, delta, and kappa opioid receptors. Biochemistry 40 (2001) 8018-8029.

[44] E. Turk, C.J. Kerner, M.P. Lostao, and E.M. Wright, Membrane topology of the human Na+/glucose cotransporter SGLT1, J Biol Chem 271 (1996) 1925-1934.

[45] E.R. Bennett, and B.l. Kanner, The membrane topology of GAT-1, a (Na+ + CI-)-coupled gamma- amlnobutyric acid transporter from rat brain, J Biol Chem 272 (1997) 1203-1210.

[46] W.L. Hubbell, A. Gross, R. Langen, and M.A. Lletzow, Recent advances In site-directed spin labeling of proteins, Curr Opin Struct Biol 8 (1998) 649-656.

[47] W.L. Hubbell, D.S. Caflso, and C. Altenbach, Identifying conformational changes with site-directed spin labeling, Nat Struct Biol 7 (2000) 735-739.

[48] M. Barranger-Mathys, and D.S. Caflso, Membrane structure of voltage-gated channel forming peptides by site- directed spin-labeling. Biochemistry 35 (1996) 498-505.

132 PrSompompi8trtranchErPefozorStrueturfrof-theKc»A- channel intracellular gate in the open state, Nat Struct Biol 8 (2001) 883-887.

[50] C. Wegener, S. Tebbe, H.J. Steinhoff, and H. Jung, Spin labeling analysis of structure and dynamics of the Na(+)/proiine transporter of Escherichia coli. Biochemistry 39 (2000) 4831-4837.

[51] K. Jung, H. Jung, J. Wu, G.G. Prive, and H.R. Kaback, Use of site- directed fluorescence labeling to study proximity relationships in the lactose permease of Escherichia coil. Biochemistry 32 (1993) 12273-12278.

[52] K. Jung, H. Jung, and H.R. Kaback, Dynamics of lactose permease of Escherichia coli determined by site- directed fluorescence labeling. Biochemistry 33 (1994) 3980-3985.

[53] M. Zhao, K.C. Zen, W.L. Hubbell, and H.R. Kaback, Proximity between Glu126 and Arg144 in the lactose permease of Escherichia coil. Biochemistry 38 (1999) 7407-7412.

[54] R.C. Hresko, H. Murata, B.A. Marshall, and M. Mueckler, Discrete structural domains determine differential endoplasmic reticulum to Golgi transit times for glucose transporter isoforms, J Biol Chem 269 (1994) 32110-32119.

[55] K. Nagai, and H.C. Thogersen, Generation of beta-globin by sequence-specific proteolysis of a hybrid protein produced in Escherichia coli. Nature309 ft984) B10-at2.

[56] K. Bamberg, and G. Sachs, Topological analysis of H+,K(+)-ATPase using in vitro translation, J Biol Chem 269 (1994) 16909-16919.

[57] S. Hallen, M. Branden, P.A. Dawson, and G. Sachs, Membrane insertion scanning of the human ileal sodium/bile acid co- transporter. Biochemistry 38 (1999) 11379-11388.

133 „^..™„j58hMrGhenran€l-JrFrZhangrMembrane-inserttonrPfoee9sing-,and- topology of cystio fibrosis transmembrane conductance regulator (CFTR) in microsomal membranes. Mol Membr Biol 13 (1996) 33-40.

[59] R.F. Doolittle, M.S. Johnson, I. Husain, B. Van Houten, D.C. Thomas, and A. Sancar, Domainal evolution of a prokaryotic DNA repair protein and its relationship to active-transport proteins. Nature 323 (1986) 451-453.

[60] D.S. Dwyer, Model of the 3-D structure of the GLUT3 glucose transporter and molecular dynamics simulation of glucose transport, Proteins 42 (2001) 531-541.

[61] T. Hiroike, J. Higo, H. Jingami, and H. Toh, Homology modeling of human leptin/leptin receptor complex, Biochem Biophys Res Commun 275 (2000) 154-158.

[62] J.M. Wurtz, B. Guillot, J. Fagart, D. Moras, K. Tietjen, and M. Schindler, A new model for 20-hydroxyecdysone and dibenzoylhydrazine binding: a homology modeling and docking approach, Protein Sci 9 (2000) 1073- 1084.

[63] D. Halaby, E. Thoreau, J. Djiane, and J.P. Mornon, Homology modeling of rabbit prolactin hormone complexed with its receptor, Proteins 27 (1997) 459-468.

[64} G. Costantino, and R. Pellicciarr, Homology modeling of metabotropic glutamate receptors. (mGluRs) structural motifs affecting binding modes and pharmacological profile of mGluRI agonists and competitive antagonists, J Med Chem 39 (1996) 3998-4006.

[65] K.H. Kim, 3D-QSAR analysis of 2,4,5- and 2,3,4,5-substituted imidazoles as potent and nontoxic modulators of P-glycoprotein mediated MDR, Bioorg Med Chem 9 (2001) 1517-1523.

134 —— —[68}-MijrBorgmiarG:DrEyt@n, and“Y:G. AsaarafrGompelHioiTof hydrophobic peptides, cytotoxic drugs, and chemosensitizers on a common P* glycoprotein pharmacophore as revealed by its ATPase activity, J Biol Chem 271 (1996)3163-3171.

[67] P.W. Swaan, and J.J. Tukker, Molecular determinants of recognition for the intestinal peptide carrier, J Pharm Sci 86 (1997) 596-602.

[68] P.W. Swaan, P.O. Szoka, Jr., and S. Oie, Moiecular modeling of the intestinal bile acid carrier: a comparative molecular field analysis study, J Comput Aided Moi Des 11 (1997) 581-588.

[69] K.H. Baringhaus, H. Matter, S. Stengelin, and W. Kramer, Substrate specificity of the ileal and the hepatic Na(+)/bile acid cotransporters of the rabbit. II. A reliable 3D OSAR pharmacophore model for the ileal Na(+)/bile acid cotransporter, J Lipid Res 40 (1999) 2158-2168.

[70] S.J. Oh, H.J. Ha, D.Y. Chi, and H.K. Lee, Serotonin receptor and transporter ligands - current status, Curr Med Chem 8 (2001) 999-1034.

[71] S. Wang, S. Sakamuri, I.J. Enyedy, A.P. Kozikowski, 0 . Deschaux, B.C. Bandyopadhyay, S.R. Telia, W.A. Zaman, and K.M. Johnson, Discovery of a novel dopamine transporter inhibitor, 4-hydroxy-1-methyl- 4-(4-methylphenyl)-3- piperidyl 4-methylphenyl ketone, as a potential cocaine antagonist through 3D- datatiase pharmacophore searching. Molecular modeling, structure-activity relationships, and behavioral pharmacological studies, J Med Chem 43 (2000) 351-360.

[72] S. Sakamuri, I.J. Enyedy, A.P. Kozikowski, W.A. Zaman, K.M. Johnson, and S. Wang, Pharmacophore-based discovery, synthesis, and biological evaluation of 4-phenyl-1 -arylalkyl piperidines as dopamine transporter inhibitors, Bioorg Med Chem Lett 11 (2001) 495-500.

[73] P.W. Swaan, K.M. Hillgren, F.C. Szoka, Jr., and S. Oie, Enhanced transepithelial transport of peptides by conjugation to cholic acid, Bioconjug Chem 8 (1997) 520-525.

135 [74] W. Kramer, G. Wess, A. Enhsen, K. Bock, E. Falk, A. Hoffmann, G. Neckermann, D. Gantz, S. Schulz, L. Nlckau, and et al.. Bile acid derived HMG- CoA reductase inhibitors, Biochim Biophys Acta 1227 (1994) 137-154.

[75] N.N. Izzat, M.E. Oeshazer, and D.S. Loose-Mitcheil, New molecular targets for cholesterol-lowering therapy, J Pharmacol Exp Ther 293 (2000) 315- 320.

[76] G.R. Thompson, and R.P. Naoumova, Novel lipid-regulating drugs. Expert Opin Investig Drugs 9 (2000) 2619-2628.

[77] K.N. Lazaridis, P. Tietz, T. Wu, S. Kip, P.A. Dawson, and N.F. LaRusso, Alternative splicing of the rat sodium/bile acid transporter changes its cellular localization and transport properties, Proc Natl Acad Sci U S A 9 7 (2000) 11092-11097.

[78] T. Hirohashi, H. Suzuki, H. Takikawa, and Y. Sugiyama, ATP- dependent transport of bile salts by rat multidrug resistance- associated protein 3 (MrpS), J Biol Chem 275 (2000) 2905-2910.

[79] J. Noe, B. Hagenbuch, P.J. Meier, and M.V. St-Pierre, Characterization of the mouse bile salt export pump overexpressed in the baculovirus system, Hepatology 33 (2001) 1223-1231.

[80] M. Ananthanarayanan, N.V. Balasubramanian, M. Makishima, D.J. Mangelsdorf, and F.J. Suchy, Human bile salt export pump (BSEP) promoter is transactivated by the famesoid X receptor/bile acid receptor (FXR/BAR), J Biol Chem 31 (2001)31.

[81] K.N. Lazaridis, L. Pham, P. Tietz, R.A. Marinelli, P.C. deGroen, S. Levine, P.A. Dawson, and N.F. LaRusso, Rat cholangiocytes absorb bile acids at their apical domain via the ileal sodium-dependent bile acid transporter, J Clin Invest 100 (1997) 2714-2721.

136 [82] N. Chignard, M. Mergey, D. Veissiere, R. Parc, J. Capeau, R. Poupon, A. Paul, and C. Housset, Bile acid transport and regulating functions in the human biliary epithelium, Hepatology 33 (2001) 496-503.

[83] 8 . Hagenbuch, B. Stieger, M. Foguet, H. Lubbert, and P.J. Meier, Functional expression cloning and characterization of the hepatocyte Na+/bile acid cotransport system, Proc Natl Acad Sci U S A 88 (1991) 10629-10633.

[84] B. Hagenbuch, and P.J. Meier, Molecular cloning, chromosomal localization, and functional characterization of a human liver Na+/bile acid cotransporter, J Clin Invest 93 (1994) 1326-1331.

[85] M.H. Wong, P. Oelkers, A.L. Craddock, and P.A. Dawson, Expression cloning and characterization of the hamster ileal sodium- dependent bile acid transporter, J Biol Chem 269 (1994) 1340-1347.

[86] M.H. Wong. P. Oelkers, and P.A. Dawson, Identification of a mutation in the ileal sodium-dependent bile acid transporter gene that abolishes transport activity, J Biol Chem 270 (1995) 27228-27234.

[87] D.M. Christie, P.A. Dawson, S. Thevananther, and B.L. Shneider, Comparative analysis of the ontogeny of a sodium-dependent bile acid transporter in rat kidney and ileum. Am J Physiol 271 (1996) G377-385.

[88] T. Saeki, K. Matoba, H. Furukawa, K. Kirifuji, R. Kanamoto, and K. Iwami, Characterization. cDNA cloning, and functional expression of mouse ileal sodium-dependent bile acid transporter, J Biochem (Tokyo) 125 (1999) 846-851.

[89] P. Oelkers, L.C. Kirby, J.E. Heubi, and P.A. Dawson, Primary bile acid malabsorption caused by mutations in the ileal sodium- dependent bile acid transporter gene (SLC10A2), J Clin Invest 99 (1997) 1880-1887.

137 [90]-WrKfafnefrSrSteng9ljn, KrHrBaringhaus, Ar.-EnhsenrHr HetierrWr Becker, D. Corsiero, F. Girbig, R. Noli, and C. Weyland, Substrate specificity of the lieai and the hepatic Na(+)/biie acid cotransporters of the rabbit, i. Transport studies with membrane vesicles and ceil lines expressing the cloned transporters. J Lipid Res 40 (1999) 1604-1617.

[91] A. Schroeder, U. Eckhardt, B. Stieger, R. Tynes, C.D. Schteingart, A.F. Hofmann, P.J. Meier, and B. Hagenbuch, Substrate specificity of the rat liver Na(+)-biie salt cotransporter in Xenopus iaevis oocytes and in CHO cells. Am J Physiol 274 (1998) G370-375.

[92] A.L. Craddock, M.W. Love, R.W. Daniel, L.C. Kirby, H.C. Walters, M.H. Wong, and P.A. Dawson, Expression and transport properties of the human ileal and renal sodium- dependent bile acid transporter. Am J Physiol 274 (1998) G157-169.

[93] P.J. Meier, Molecular mechanisms of hepatic bile salt transport from sinusoidal blood into bile. Am J Physiol 269 (1995) G801-812.

[94] M.H. Wong, P.M. Rao, M.J. Pettenati, and P.A. Dawson, Localization of the ileal sodium-bile acid cotransporter gene (SLC10A2) to human chromosome 13q33, Genomics 33 (1996) 538-540.

[95] B.L. Shneider, P.A. Dawson, D.M. Christie, W. Hardikar, M.H. Wong, and F.J. Suchy, Cloning and molecular characterization of the ontogeny of a rat ileal sodium-dependent bile acid transporter, J Clin Invest 95 (1995) 745-754.

[96] G. Alpini, S.S. Glaser, R. Rodgers, J .L Phinizy, W.E. Robertson, J. Lasater, A. Caligiuri, Z. Tretjak, and G.D. LeSage, Functional expression of the apical Na+ dependent bile acid transporter in large but not small rat cholangiocytes. Gastroenterology 113 (1997) 1734-1740.

[97] A. Benedetti, A. Di Sario, L. Marucci, G. Svegliati-Baroni, C.D. Schteingart, H.T. Ton-Nu, and A.F. Hofmann, Carrier-mediated transport of conjugated bile acids across the basolateral membrane of biliary epithelial cells. Am J Physiol 272 (1997) G1416-1424.

138 [98] W.C. Duane, L.A. Hartich, A.E. Bartman, and S B. Ho, Diminished gene expression of ileal apical sodium bile acid transporter explains impaired absorption of bile acid in patients with hypertriglyceridemia, J Lipid Res 41 (2000) 1384-1389.

[99] M.J. Nowicki, B.L. Shneider, J.M. Paul, and J.E. Heubi, Glucocorticoids upregulate taurocholate transport by ileal brush-border membrane. Am J Physiol 273 (1997) G197-203.

[100] J.V. Higgins, J.M. Paul, R. Dumaswala, and J.E. Heubi, Downregulation of taurocholate transport by ileal BBM and liver BLM in biliary- diverted rats. Am J Physiol 267 (1994) G501-507.

[101] J. Lillienau, D.L. Crombie, J. Munoz, S.J. Longmire-Cook, L.R. Hagey, and A.F. Hofmann, Negative feedback regulation of the ileal bile acid transport system in rodents. Gastroenterology 104 (1993) 38-46.

[102] R.T. Stravitz, A.J. Sanyal, W.M. Pandak, Z.R. Vlahcevic, J.W. Beets, and P.A. Dawson, Induction of sodium-dependent bile acid transporter messenger RNA, protein, and activity in rat ileum by cholic acid. Gastroenterology 113 (1997) 1599-1608.

[103] S.A. Weinman, M.W. Carruth, and P.A. Dawson, Bile acid uptake via the human apical sodium-bile acid cotransporter is electrogenic, J Biol Chem 273 (1998) 34691-34695.

[104] E. Fayard, K. Schoonjans, and J. Auwerx, Xol INXS: role of the liver X and the farnesoi X receptors, Curr Opin Lipidol 12 (2001) 113-120.

[105] J R. Walters, Bile acids are physiological ligands for a nuclear receptor. Gut 46 (2000) 308-309.

139 — ~[4'0S}-M;-Mski8hifnsr As-¥s-Ok€WflOt©r«k*K“R0pfl, H.-Tut- RïMï'Lôôrnodv Ar Luk, M.V. Hull, K.D. Lustig, D.J. Mangelsdorf, and B. Shan, Identification of a nuclear receptor for bile acids. Science 284 (1999) 1362-1365.

[107] M.S. Tollefson, W.F. Vernier, H.C. Huang, P.P. Chen, E.J. Reinhard, J. Beaudry, B.T. Keller, and D.B. Reitz, A novel class of apical sodium co­ dependent bile acid transporter inhibitors; the 2,3-disubstituted-4- phenylquinolines, Bioorg Med Chem Lett 10 (2000) 277-279.

[108] L. Lack, and I.M. Weiner, Intestinal bile salt transport: structure- activity relationships and other properties. Am J Physiol 210 (1966) 1142-1152.

[109] L. Lack, A. Tantawi, C. Halevy, and D. Rockett, Positional requirements for anionic charge for iieal absorption of bile salt analogues. Am J Physiol 246 (1984) G745-749.

[110] T.S. Low-Beer, M.P. Tyor, and L. Lack, Effects of sulfation of taurolithochoiic and giycolithocholic acids on their intestinal transport. Gastroenterology 56 (1969) 721-726.

[ I l l ] L. Lack, J.T. Walker, and G.D. Singletary, Ileal bile salt transport: in vivo studies of effect of substrate ionization on activity. Am J Physiol 219 (1970) 487-490.

[112] E.H. De Witt, and L. Lack, Effects of sulfation patterns on intestinal transport of bile salt sulfate esters, AnvJ Physiol 23&(1980> G34-39.

[113] S. Walker, A. Stiehl, R. Raedsch, P. Kloters, and B. Kommerell, Absorption of urso- and chenodeoxycholic acid and their taurine and glycine conjugates in rat jejunum, ileum, and colon. Digestion 32 (1985) 47-52.

[114] N.F. Ho, Utilizing bile acid carrier mechanisms to enhance liver and small intestine absorption, Ann N Y Acad Sci 507 (1987) 315-329.

140 - f1-1'5J“W-.-KFafn©FrGr'W©99rGtrSchub©rtrMr Bfek6lr"F.' Girbt^ U r Gutjahr; S. Kowalewski, K.H. Baringhaus, A. Enhsen, H. Glombik, and et al.. Liver- specific drug targeting by coupling to bile acids, J Bioi Chem 267 (1992) 18598- 18604.

[116] G. Wess, W. Kramer, X.B. Han, K. Bock, A. Enhsen, H. Glombik, K.H. Baringhaus, G. Boger, M. Urmann, A. Hoffmann, and et al.. Synthesis and bioiogical activity of bile acid-derived HMG-CoA reductase inhibitors. The role of 21-methyl in recognition of HMG-CoA reductase and the iieal bile acid transport system, J Med Chem 37 (1994) 3240-3246.

[117] Z.F. Stephan, E C. Yurachek, R. Sharif, J.M. Wasvary, R E. Steele, and C. Howes, Reduction of cardiovascular and thyroxine-suppressing activities of L- T3 by liver targeting with cholic acid, Biochem Pharmacol 43 (1992) 1969- 1974.

[118] D.C. Kim, A.W. Harrison, M.J. Ruwart, K.F. Wilkinson, J.F. Fisher, I.J. Hidalgo, and R.T. Borchardt, Evaluation of the bile acid transporter in enhancing intestinal permeability to renin-inhibitory peptides, J Drug Target 1 (1993) 347-359.

[119] C.O. Mills, G.H. Martin, and E. Elias, The effect of tyrosine conjugation on the criticai micellar concentration of free and glycine-conjugated bile salts, Biochim Biophys Acta 876 (1986) 677-683.

[120] M.S. Anwer, E.R. O’Maiile, A.F. Hofmann, R.A. DiPietro, and E. Michelotti, Influence of side-chain charge on hepatic transport of bile acids and bile acid analogues. Am J Physiol 249 (1985) G479-488.

[121] C.O. Mills, S. Iqbal, and E. Elias, Selectively reduced biliary excretion of choiyldiglycylhistamine but not of cholyitetraglycyihistamine in ethinyl estradiol-treated rats. A possible indicator of increased bile canalicular permeability, J Hepatol 1 (1985) 199-210.

[122] M-. Ast, and W.H. Frishman, Bile acid séquestrants, J Clin Pharmacol 30 (1990) 99-106.

141 [123] G. Wess, W. Kramer, A. Enhsen, H. Glombik, K.H. Baringhaus, G. Boger, M. Urmann, K. Bock, H. Kleine, G. Neckermann, and et al.. Specific inhibitors of ileal bile acid transport, J Med Chem 37 (1994) 873-875.

[124] C. Root, C.D. Smith, D.A. Winegar, LE . Brieaddy, and M.C. Lewis, Inhibition of ileal sodium-dependent biie acid transport by 2164U90, J Lipid Res 36(1995) 1106-1115.

[125] M.C. Lewis, L.E. Brieaddy, and C. Root, Effects of 2164U90 on ileal bile acid absorption and serum cholesterol in rats and mice, J Lipid Res 36 (1995) 1098-1105.

[126] T. Yamaguchi, Y. Nakajima, M. Mizobuchi, K. Inazawa, T. Kanazu, K. Kadono, T. Ohkawa, and K. Iwatani, Disposition and metabolism of the new hypocholesterolemic compound S- 8921 in rats and dogs, Arzneimittelforschung 48 (1998) 995-1006.

[127] J. Higaki, S. Mara, N. Takasu, K. Tonda, K. Miyata, T. Shike, K. Nagata, and T. Mizui, Inhibition of ileal Na+/bile acid cotransporter by S-8921 reduces serum cholesterol and prevents atherosclerosis in rabbits, Arterioscler Thromb Vase Biol 18(1998) 1304-1311.

[128] G. Kominami, A. Ueda, K. Sakai, and A. Misaki, Radioimmunoassay for a novel lignan-related hypocholesterolemic agent, S-8921, in human plasma after high-performance liquid chromatography purification and in human urine after immunoaffmity extraction, J Chromatogr B Biomed Sci AppI 704 (1997) 243- 250.

[129] T. Ichihashi, M. Izawa, K. Miyata, T. Mizui, K. Hirano, and Y. Takagishi, Mechanism of hypocholesterolemic action of S-8921 in rats; S-8921 inhibits ileal bile acid absorption, J Pharmacol Exp Ther 284 (1998) 43-50.

[130] S. Mara, J. Higaki, K. Higashino, M. Iwai, N. Takasu, K. Miyata, K. Tonda. K. Nagata, Y. Goh, and T. Mizui, S-8921, an ileal Na+/bile acid

142 -eotFaFi8peFte^inhibitoFdeeFea8es-8eiwr^ehole8teFel-in-ham9tef&, yfe-Sch60- (1997) 365-370.

[131] R.J. Bahar, and A. Stolz, Bile acid transport, Gastroenterol Clin North Am 28 (1999) 27-58.

[132] P.A. Dawson, and P. Oelkers, Bile acid transporters, Curr Opin Lipidol 6 (1995) 109-114.

[133] M.V. St-Pierre, G.A. Kullak-Ublick, B. Hagenbuch, and P.J. Meier, Transport of bile acids in hepatic and non-hepatic tissues, J Exp Biol 204 (2001) 1673-1686.

[134] D. Bayle, D. Weeks, S. Malien, K. Melchers, K. Bamberg, and G. Sachs, In vitro translation analysis of integral membrane proteins, J Recept Signal Transduct Res 17 (1997) 29-56.

[135] K.J. Buck, and S.G. Amara, Chimeric dopamine-norepinephrine transporters delineate structural domains influencing selectivity for catecholamines and 1-methyl-4- phenylpyridinium, Proc Natl Acad Sci U S A 91 (1994) 12584-12588.

[136] I.M. Nilsson, and G. von Heijne, Determination of the distance between the oligosaccharyltransferase active site and the endoplasmic reticulum membrane, J Biol Chem 268 (1993) 5798-5801.

[137] A.Q. Sun, M.A. Arrese, L. Zeng, I. Swaby, M.M. Zhou, and F.J. Suchy, The rat liver Na(+)/bile acid cotransporter. Importance of the cytoplasmic tail to function and plasma membrane targeting, J Biol Chem 276 (2001) 6825- 6833.

[138] A.Q. Sun, M. Ananthanarayanan, C.J. Soroka, S. Thevananther, B.L. Shneider, and F.J. Suchy, Sorting of rat liver and ileal sodium-dependent

143 bile^acid-transporter8~irwpolarizecl-epitiielial-eeil&rAm^J-Phy8jel^27&ft99S> G1045-1055.

[139] S. Hall?en, J. Fryklund, and G. Sachs, Inhibition of the human sodium/biie acid ootransporters by side- specific methanethiosuifonate sulfhydryl reagents: substrate-controlled accessibility of site of inactivation, Biochemistry 39 (2000) 6743-6750.

[140] W. Kramer, F. Girbig, H. Glombik, D. Corsiero, S. Stengelin, and C. Weyland, Identification of a ligand binding site in the Na+/bile acid cotransporting protein from rabbit ileum, J Biol Chem 10 (2001) 10.

[141] B. Hagenbuch, Molecular properties of hepatic uptake systems for bile acids and organic anions, J Membr Biol 160 (1997) 1-8.

[142] W. Kramer, S B. Nicol, F. Girbig, U. Gutjahr, S. Kowalewski, and H. Fasold, Characterization and chemicai modification of the Na(+)-dependent bile- acid transport system in brush-border membrane vesicles from rabbit ileum, Biochim Biophys Acta 1111 (1992) 93-102.

[143] K.M. Penado, G. Rudnick, and M.M. Stephan, Critical amino acid residues in transmembrane span 7 of the serotonin transporter identified by random mutagenesis, J Biol Chem 273 (1998) 28098-28106.

[144] J. Wu, and H R. Kaback, A general method for determining heiix packing in membrane proteins in situ: heiicee I ancLILare close to helix VU in the lactose permease of Escherichia coli, Proc Natl Acad Sci U S A93 (1996) 14498- 14502.

[145] J. Wu, D.M. Perrin, D.S. Sigman, and H R. Kaback, Helix packing of lactose permease in Escherichia coil studied by site- directed chemicai cleavage, Proc Natl Acad Sci U S A 92 (1995) 9186-9190.

144 “f t4ehYrKirmirarB:GrVas8ylyevrArMiya2awarA: Kiderar M; Matsoshimar K. Mitsuoka, K. Murata, T. Mirai, and Y. Fujiyoshi, Surface of bacteriorfiodopsin revealed by high-resolution electron crystallography, Nature 389 (1997) 206*211.

[147] D. Fu, A. Libson, L.J. Miercke, C. Weitzman, P. Nollert, J. Krucinski, and R.M. Stroud, Structure of a glycerol-conducting channel and the basis for its selectivity. Science 290 (2000) 481*488.

[148] G. Chang, and C.B. Roth, Structure of MsbA from E. coli: a homolog of the multidrug resistance ATP binding cassette (ABC) transporters. Science 293 (2001) 1793-1800.

[149] Y.Z. Gong, E.T. Everett, D.A. Schwartz, J.S. Norris, and F A. Wilson, Molecular cloning, tissue distribution, and expression of a 14-kDa bile acid- binding protein from rat ileal cytosol, Proc Natl Acad Sci U S A91 (1994) 4741* 4745.

[150] P. Oelkers, and P.A. Dawson, Cloning and chromosomal localization of the human ileal lipid-binding protein, Biochim Biophys Acta 1257 (1995) 199* 202.

[151] S. Stengelin, S. Apel, W. Becker, M. Maier, J. Rosenberger, U. Bewersdorf, F. Girbig, C. Weyland, G. Wess, and W. Kramer, The rabbit ileal lipid-binding protein. Gene cloning and functional expression of the recombinant protein, Eur J Biochem 239 (1996) 887-896.

[152] W. Kramer, F. Girbig, U. Gutjahr, S. Kowalewski, K. Jouvenal, G. Muller, D. Tripier, and G. Wess, Intestinal bile acid absorption. Na(+)-dependent bile acid transport activity in rabbit small intestine correlates with the coexpression of an integral 93-kDa and a peripheral 14-kDa bile acid-binding membrane protein along the duodenum-ileum axis, J Biol Chem 268 (1993) 18035*18046.

[153] W. Kramer, G. Wess, U. Bewersdorf, D. Corsiero, F. Girbig, C. Weyland, S. Stengelin, A. Enhsen, K. Bock, H. Kleine, M.A. Le Dreau, and H.L.

145 “Schafer, TopologicaKphotoaff*^hi(y ^l@beNng-o^the^rabbi^HeakNa+/bile-saH- cotransport system, Eur J Biochem 249 (1997) 456-464.

[154] W. Kramer, F. Girbig, U. Gutjahr, and S. Kowalewski, Radiation- inactivation analysis of the Na+/blle acid co-transport system from rabbit ileum, Biochem J 306 (1995) 241-246.

[155] W. Kramer, D. Corsiero, M. Friedrich, F. Girbig, 8 . Stengelin, and 0 . Weyland, Intestinal absorption of bile acids: paradoxical behaviour of the 14 kOa ileal lipld-binding protein in differential photoaffinity labelling, Biochem J 333 (1998) 335-341.

[156] W. Kramer, K. Sauber, K.H. Baringhaus, M. Kurz, S. Stengelin, G. Lange, D. Corsiero, F. Girbig, W. Konig, and C. Weyland, Identification of the bile acid binding site of the ileal llpid binding protein (ILBP) by phoatoaffinity labeling, MALDI mass spectrometry and NMR structure, J Biol Chem 7 (2000) 7.

[157] E.M. Phlzlcky, and S. Fields, Proteln-protein interactions: methods for detection and analysis, Microbiol Rev 59 (1995) 94-123.

[158] A.R. Mendelsohn, and R. Brent, Protein interaction methods-toward an endgame. Science 284 (1999) 1948-1950.

[159] B. Kuster, and M. Mann, Identifying proteins and post-translational modifications by mass spectrometry, Curr Opin Struct Biol 8 (1998) 393-400.

[160] G. RIgaut, A. Shevchenko, B. Rutz, M. Wllm, M. Mann, and B. Séraphin, A generic protein purification method for protein complex characterization and proteome exploration, Nat Biotechnol 17 (1999) 1030-1032.

[161] J R. Yates, 3rd, Mass spectrometry. From genomics to proteomics. Trends Genet 16 (2000) 5-8.

146

: j. "-ft6g)-Ar-Attkenr Protejn consensuy sequence motifs, MohBiotechnol t2^ (1999) 241-253.

[163] A.J. Enright, I. liiopoulos. N.C. Kyrpides, and C.A. Ouzounis, Protein interaction maps for complete genomes based on gene fusion events, Nature 402 (1999) 86-90.

[164] E.M. Marcotte, M. Pellegrini, H.L. Ng, D.W. Rice, T.O. Yeates, and D. Eisenberg, Detecting protein function and protein-protein interactions from genome sequences, Science 285 (1999) 751-753.

[165] M. Hubsman, G. Yudkovsky, and A. Aronheim, A novel approach for the identification of protein-protein interaction with integral membrane proteins. Nucleic Acids Res 29 (2001) E l 8 .

147 APPENDIX

148 AAPS Phamud 1999; 1 (4) artiel* 20 (httpr/Awww.phannsci.org/)

Determination of Membrane Protein Glycation In Diabetic Tissue

SubndBd:SefitBmter7,1999; Accepted:December21,1999; Pubfshed:December30,1999

Eric Y. Zhang and Peter W . Swaan Division of PhamnaceuOcs. Coteoe of Phatmacv. The Ohio State Üntversitv. Cotumbus. OH 43210-1291 ABSTRACT Diabetes-associated hyperglycemia Introduction causes glycation o f proteins at reactive amino Hyperglycemia, the clinical hallmaric o f poorly groups, which can adversely affect protein function. controlled diabetes, is known to cause protein Although the effects o f glycation on soluble proteins glycation, also known as nonenzymic glycosylation are well characterized, there is no infonnation (1). Protein glycation takes place when elevated regarding membrane-associated proteins, mainly levels o f circulating reduced sugars react with amino because o f the lack o f reproducible methods to groups in proteins, forming a labile aldimine Schiff determine protein glycation in vivo. The current base, which rearranges to form a stable oxoamine study was conducted to establish such a method and sugar adduct known as an Amadori product. The to compare the glycation levels o f membrane- reactive amino group can be either an N-terminal a- associated proteins derived from normal and diabetic NH 2 or a lysyl e-NHz group, depending on its tissue. We present a detailed sample preparation accessibility and environment. Glycation has been protocol based on the borohydride-periodate assay, found to occur both in vitro and in vivo. In fact, this modified to allow manipulation o f animal tissue. reaction is responsible for the formation o f glycated Assay noise associated with extraction protocols and hemoglobin (GHb) during diabetic hyperglycemia, nonproteinaceous buffer components was eliminated and the quantification o f this Amadori product is by the using 3-[(3-cholamidopropyl) used clinically as an index of diabetic control. Other dimethylammonio] -1 -propanesulAnate (CHAJ*S) as long-lived proteins have been found to undergo a membrane detergent, applying desalting columns, glycation in vivo, such as lens crystallins, collagen, and including a protein precipitation step. The low-density lipoprotein, albumin, and fibronectin (2). glycation level of membrane proteins from diabetic rats is elevated to 4.89 nmol/mg protein (standard Glycation can drastically affect protein function, deviation [SD] 0.48) compared with normoglycemic exemplified by the increased activity o f the enzyme control tissue (2.23 nmol/mg protein, SD 0.64). This ribonuclease A upon glycation o f its functional result is consistent with and correlated to the total lysine residues (3). Furthermore, glycation o f lens glycated hemoglobin levels in diabetic and crystallin induces a conformational change in the normoglycemic rats. Using <100 pg protein, the protein that increases SH-group susceptibility to described methods allow further study o f protein oxidation (4). glycatToh effects on the function o f mdTviduaT Nutrient transporters are proteins with 7-14 transporter proteins and the role o f these membrane-spaiming domains and are, therefore, modifications in diabetes.______intrinsically bound to the cell membrane. They KEYWORDS: Glycation, Membrane Protein, Diabetes. perform an indispensable function in the uptake and excretion o f nutrients and cellular metabolites in ABBREVIATIONS: CHAPS, 3-[(3-cholanoidopropyl) virtually all cells throughout the human body. dimethylammonio]-l-propanesulfonate; GHb, glycated Furthermore, it has been recognized that nutrient hemoglobin; HH*ES, 4(-2-hydroxyethyl)-l- transporters play an invaluable role in overall drug piperazineethanesulfbnic acid; PBS, phosphate- absorption, distribution, metabolism, and excretion. buffered saline; PMSF, phenylmethylsulfonyl fluoride; Since nutrient transporters are intrinsic membrane SOS, sodium dodecyl sulfate. proteins, their transmembrane protein regions w ill likely have a limited exposure to reducing sugars, but CenMpendbta Auttion Peter W. Swaan. College of Phaimacy, their extracellular domains w ill not be shielded. 500 West 12 Ave, Columbus, OH; telephone: (614) 688-5609; Bilan and Klip (2) have previously shown glycation ISceimile: (614) 292-7766; e-mail: [email protected]. AAPS Phtmucl 1999; 1 (4) artiel* 20 (http://Www.phannsci.org/)

ÀnSiârTreatmettt isolated erythrocytes in vitro, but data on transporter Diabetes was induced in Sprague-Dawley rats by glycation in tissues in vivo are not available, most intraperitoneal injection o f 60 mg/kg streptozotocin likely because o f the lack o f reproducible methods as described previously (8). The rats were killed 6 for measuring membrane protein glycation in vivo. weeks after injection when plasma glucose levels The need for a protocol to assess hyperglycemia- were >500 mg/dL. Total GHb in control and diabetic associated modifications o f membrane transporters rats was measured using a GHb assay kit following during the course o f diabetes prompted the work the manufacturer’s protocol. This assay uses an described in this article. affinity chromatography procedure for specifically Among the existing protein glycation assays, isolating glycated hemoglobins as described by quantitation o f the Amadori product via periodate Oremeketal. (9). oxidation remains the most convenient. This method Extraction o f Membrane Proteins From Isolated was first proposed by Gallop et al. (5) to quantify the Crude Membrane Fragments glycation level of hemo^obin and was further adapted by using a microplate reader(6) with The following procedure was used to specifically sensitivity enhancement (0.1 mg protein). Another isolate the membrane protein firaction fk m other important improvement was achieved by reducing lipid-containing cellular fractions, such as subcellular proteins with sodium borohydride before periodate membranes (mitochondria, peroxisomes, lysosomes, oxidation, allowing the successful measurement o f or /rans-Golgi network) or the cell nucleus. Whole tire glycation level o f serum glycoproteins, which are intestinal mucosa was removed fiom diabetic and heavily glycosylated and thus contain terminal normoglycemic (control) rats. The crude membrane reducing sugars (7 ) that would normally cause fiactions were isolated using the method described excessive background noise in the typical periodate by Marshak (10), with slight modifications. In brief, assay. However, when we applied this method to the mucosa was minced in ice-cold homogenization membrane proteins extracted fiom the intestine o f buffer (250 mM sucrose, 10 mM 4(-2-hydroxyethyl)- diabetic and normoglycemic rats, we found the assay l-piperazineethanesulfonic acid [HEPES], 1 mM to be highly sensitive to nonproteinaceous ethylenediaminetetraacetic acid [EDTA] and I mM endogenous substances. Furthermore, the method phenylmethylsulfonyl fluoride [PMSF]) and appeared to be sensitive to buffer components used transferred to a Bounce homogenizer. After 15 in conunon membrane protein extraction and strokes with pestle A, the homogenate was isolation protocols. Here, we describe a strategy to centrifuged at 8,000g (10 minutes, 4°C, Beckman overcome technical difficulties that may serve as a model J2-21). The pellet was discarded, and the general way to determine the glycation level o f supernatant was centrifuged at I00,000g for 20 membrane proteins extracted fiom tissues by using minutes at 4°C in a Beckman ultracentrifUge (model ~the~ bdrohydride-peribdate assay. This technique L5-7Sy. Subsequently, the supernatant was discarded should be o f particular importance for investigating and the pellet was resuspended in homogenization the role o f glycation o f membrane proteins in the buffer, homogenized using a Bounce homogenizer, pathogenesis o f diabetes. and ultracentrifuged using the protocol described above. The supernatant was diseased, and the pellet was mixed thoroughly and incubated with l.O mL MATERIALS An d Metho ds phosphate-buffered saline (PBS) containing 0.5% 3- Chemicals [(3-cholamidopropyl) dimethylammonio]-1- A GHb assay kit was purchased from Sigma (St. propanesulfbnate (CHAPS) and I mM PMSF for 30 Louis, M O ). Chemicals were obtained fiom minutes at 4°C. After the next centrifugation step Acros/Fisher Scientific and Sigma and were used (10,000 rpm, 30 minutes, 4°C, using a micromax RF without further purification. microcentrifiige [lEC], which was used for all subsequent centrifugation steps), the pellet was discarded, and the supernatant was collected. The AAPS Phamuel 1M0; 1 (4) aitiel* 20 (http:/Awww.pharmMi.erg/)

protein extract and was used for the subsequent detection system. sample preparation steps. Borohydride-Periodate Assay Sample Preparation A previously described borohydride-periodate assay The fraction containing total membrane fiagments, (7) was used with several modifications. Triplicate collected via the procedure described above, was aliquots o f each sample (100 pL) were incubated for adjusted to 2.5 mL in volume with extraction buffer 1 hour at room temperature with 20 pL o f 0.2 M and loaded onto a PD-10 column (Pharmacia, which sodium borohydride in ice-cold 0.01 mM NaOH, was pre-equilibrated with PBS containing 0.2% using 20 pL o f 0.01 mM NaOH as a control CHAPS. The first 3 mL o f eluant was collected, and (triplicates). This step ensures reduction o f protein 500 pL o f this fraction was transferred to a new 1.5 glycosylation residues before conversion o f glycation mL microcentrifuge tube. Subsequently, 400 pL residues by periodate oxidation. The reduction distilled water and 100 pL o f 72% trichloroacetic reaction was stopped by adding 20 pL o f 0.2 M HCl. acid (TCA) solution were added. After mixing After this step, each sample was incubated with 20 thoroughly, the protein precipitate was obtained by pL o f 0.1 M sodium periodate for 30 minutes at centrifugation at 10,000 rpm for 10 minutes at 4°C. room temperature. The oxidation was terminated by The protein pellet was washed twice with 1 mL cooling the samples on ice for 10 minutes, followed water and redissolved in 1 mL of 0.1 M sodium by adding 40 pL each o f ice-cold 0.7 M NaOH and phosphate buffer containing 1% sodium dodecyl 15% zinc sulfate water solution. Precipitate was sulfate (SDS) (pH 8.0). The protein concentration in removed by two consecutive 10-minute the redissolved sample was quantified using the centrifugations at 13,000 rpm. From each sample, bicinchoninic acid method (11), and SDS- 200 p L supernatant was transferred to a polyacrylamide gel electrophoresis (SDS-PAGE) microcentrifiige tube, and 100 pL color reagent (92 was performed on 10% acrylamide gels in a Gibco* pL acetylacetone in 10 mL o f 6.6 M ammonium BRL vertical gel apparatus under reducing acetate) was added to each tube. The mixtures were conditions using the Laemmli (12) buffer system. incubated at 37°C for 1 hour. During the incubation, Gels were stained with Coomassie brilliant blue R- the presence o f excess SDS induced flocculation, 250, and separate gels were run for subsequent which can be overcome by a brief high-speed spin, immunoblotting procedures. Six 100 p L aliquots of e.g., 10 minutes at 15,000 rpm. Finally, 270 pL this final sample with a protein concentration o f 1-2 supernatant from each tube was transferred to a mg/mL were used for the borohydride-periodate regular 96-well microplate, and the detection was assay. accomplished at 405 nm in microplate reader Immunoblotting (D YN EX). Fructose solution (0-;40 nmol) was used for calibrating the periodate assay as described SDS^PAGE gels were subjected to Western blotting previously (7). Redissolving buffer (100 pL) was according to the method o f Towbin et al. (13), and incorporated in both reduction and oxidation the proteins on the gel were transferred to a reactions as blank solution. polyvinylidine difluoride (PVDF) membrane, hnmunoblotting was carried out using a rabbit Assay Validation polyclonal antibody directed against the carboxy- The current protocol was validated according to the terminal 14 amino acids o f the hamster ileal Na^-bile methods o f Kennedy et al. (7), using mixtures o f acid cotransporter (kindly provided by Dr. Paul native and in vitro glycated fetal calf fetuin as Dawson, Bowman Gray School o f Medicine, Wake internal standard. Fetuin, a highly glycosylated 48-kd Forest University, Winston-Salem, NC). Shneider et protein, is an appropriate standard kcause ~20% o f al. (14) have previously shown that this polyclonal its molecular weight comprises enzymatically linked antibody exerts cross-reactivity to detect the rat sugars, with many terminal sialic acids. Briefly, fetal transport protein. The rabbit antibody was visualized calf fetuin (5 g/L) was incubated with 0.5 M ^ucose using a horseradish peroxidase-;conjugated goat anti­ in PBS (pH 7.4) for 10 days at 37“C. Before AAPS Phënntel 1999; 1 (4) artiel* 20 (htlp:/Awww.plwniwei.efflO

-incubatioiv—bttfifef—«ohitions were stenlized—t filtration (0.22 pm filter) and contained 6 mM sodium azide to prevent microbial growth. Free sugar was removed (<10 pM ) by extensive dialysis against distilled water containing 6 ,m M N ^ 3. During dialysis, the labile Schiff base dissociates and is removed, leaving only Amadori products bound to the protein. Since fetuin is a soluble protein, the protein was spiked into tissue samples after the membrane protein extraction step. TTie accuracy o f the method was demonstrated by the ability o f the assay to give consistent measurements on a series o f protein samples containing glycated fetuin (0 ,2 5 , SO, 75, and 100%) mixed proportionally with unglycated fetuin. To further verify the accuracy o f the method, we spiked an appropriate amount o f glycated fetuin with loiown glycation level into the membrane protein F Ig in 1. Proc9dim ami taehiiiqim fw d9l9iininiiig extract and determined the glycation level o f this membrane protaingtycatkxilavtls in (fiabitic tissue. protein mixture. ' f ■ ' t. 5 The precision o f the assay was assessed by within- run and between-run validations in which we analyzed the same protein sample extracted from diabetic intestine within the same day or over 3 consecutive days, respectively. RESULTS Three diabetic rats with plasma glucose levels >500 mg/dL and three healthy (normoglycemic) age- matched control rats were killed, and their small intestine tissues were isolated. Enterocytes were collected, and the apical membrane fiactions were isolated using an isolation and sample preparation protocol illustrated in Figure 1. To ensure equal protein loading and composition, samples fiom both diabetic and control tissues were subjected to SDS-PAGE, followed by Coomassie blue staining. Representative data obtained from healthy and hyperglycemic samples are shown in Figure 2. Figure 2. Coom m ie MÜe>:italnad 10% SDS-PAGE Gel electrophoresis results (Figure 2) indicate that of iwissolirad protein samplee darivid frain dtabaiic redissolved protein samples from diabetic and and normogtycemic rat intestine. Lane 1 shows the normoglycemic rats exhibit similar molecular weight moiectiiar weight standard (lOj-gOO kd). Lanes 2 and protein distribution patterns. To ascertain that and 3 show the sampies from two normogtycemk these isolated protein samples uniquely comprise rets, ija m 4 and 5 show inainbrane proWn sarnpisa membrane proteins. Western blotting was carried out extracted from hyperglycemic tissue. Lanes 6 and 7 using 0.2 pg/mL of a specific rabbit polyclonal (InsaQ show the results of Immunoetalning for the ileal bla add transporter (48 kd) in control and diabalie Oasua^ roapectivoly. AAPS Phamuel 1999; 1 (4) artiel* 20 (http:/Awww.phann«ci.or80

-antibedy-taised agamst AecaHx«y4eaninaKM^ammo— acids o f the ileal Na*-bile acid cotransporter (IS ). Bands o f similar electrophoretic mobility (48 kd) were #9 detected in both normo^ycemic and diabetic samples using a chemiluminescence detection system. Previous studies have shown that the 48-kd band corresponds to the fully glycosylated interscapular brown adipose tissue (IB A l) protein (16). Diabetic #4 tissues show a ~2.5-fold decrease in IBAT protein expression at equal protein loading densities (calculated by average pixel density measurements o f protein bands using the program M H Image, version 1.61). The possibility exists, however, that the affinity 9 10 M S and specificity of our antibody are reduced because o f glycation o f the lysine residue in the 14-;amino acid Figure 3. Calibration graph showing the carboxy terminus it was originally raised against. We spectrophotometric absorbance at 405 nm against are currently investigating the (potential) difference in the amount (nmol) of fructose par well. Linear binding af& tity o f our antibody between unglycated rsgreeeion of the standard curve yieide a straight and synthetically glycated peptide. Regardless, these line wHh an intercept of 0.0052 nmol fhictosehviil results show the presence o f a representative nutrient and a elope of 0.0231 absorbance units per nmol transport protein in our isolated membrane protein fructose. The regression coefficient Rie 0.9998. fiaction. membrane protein extract spiked with a known A calibration graph for the periodate method was amount o f glycated fetuin. The determined glycation prepared using a fructose standard Q^igure 3). level o f the protein mixture is not significantly Fructose resembles the Amadori product form o f a different from the theoretical value (data not shown;P glycated protein in having a C-2 carbonyl group in the <0.05). straight chain configuration but existing Increased glycation o f several soluble proteins has predominantly in the ring form. Like the Amadori been observed in diabetic individuals with attendant product, it liberates 1 mole o f HCHO per mole on hyperglycemia. Normoglycemic individuals usually periodate oxidation and, therefore, can be used as a display 4%-;6% glycation o f total hemoglobin, standard fi)r the Amadori product. Figure 3 shows that whereas diabetic patients with poor glycémie control absorbance is linear in the range o f 0-;S0 nmol o f have hemoglobin glycation values o f 10% or more, a fiuctose per well, and amounts o f 2 nmol can be 2-fold increase over healthy individuals. In rats, we readily detected at absorbance values o f O.OS. At found that the level o f total GHb in diabetic (n = 6) -lower concentrations, the^ detection lim it is^ partly^ anfr normoglycemic (n = 6 ) subjects was 16.6%^ (SD prescribed by the short light path and the yield o f 0.1) and 5.2% (SD 0.1), respectively, which is in good HCHO fiom fructose. agreement with the dam reported by Junod et al. (8). In vitro glycated fetuin yielded a glycation level o f The glycation extent o f total membrane protein fitom 2.04 ± 0.08 glycation sites per molecule. This number the intestine o f diabetic rats was 4.89 nmol/mg protein closely matches the values measured by Kennedy et (SD 0.48), whereas the corresponding value for al. (7) using the original borohydride-periodate assay normoglycemic rats was 2.23 nmol/mg protein (SD (number o f glycation sites: 1.98 ± 0.07/molecule). The 0.64). The elevated glycation level o f manbrane current method also gives comparable and linear proteins from diabetic rats was consistent with their measurements on a series o f protein mixtures plasma GHb levels (Figure 4). The percent variance o f containing both glycated and nonglycated fetuin in the assay between days and among different subjects varying proportions (data not shown). Further was 4.5%. The within-run and between-run confirmation o f the accuracy o f the method was coefficients o f variance were 3.9% and 4.5%, achieved by determining the glycation level o f respectively. AAPS Phwrnnel 1999; 1 (4) artieto 20 (http://www.pharmsei.erg/)

straightfbrward-dWysis step that-was- sttfficient-fer— - sample preparation. In contrast, assessing the glycation level o f membrane proteins requires multiple sample preparation procedures: isolation o f membrane fragments, extraction o f membrane protein, and additional purification protocols. Any o f these steps may introduce new substances or faU to extract endogenous substances that can either increase background levels or interfere with the detection method itself. Unique identification o f all potentially interfering sources is virtually impossible given the complexity o f biological systems, but three distinct groups can be recognized. First, many control diabolic endogenous substances, such as cellular free glucose and other reducing sugars, can interfere with the Figura 4. The glycation iev9i (nmoi/mg protein) of total borohydride-periodate detection method. Second, the membrane protein (blue bars) extracted from diabetic hydrophobic nature o f membrane proteins requires and normoglycemic rat intestine and the the presence o f detergent throughout sample corresponding total GHb levels (% GHb, white bars) in preparation. However, we found that most nonionic their red blood ceils. {P < 0.01, unpaired t test, n > 6, detergents for recovery o f membrane components error bar» SD). Standard deviations on white bars may under nondenaturing conditions, such as the widely not be visible. used Triton X 100, dramatically decrease assay sensitivity. Finally, some commonly used buffer DISCUSSION components in membrane extraction and isolation protocols. Tris and HEPES, can cause excessive The present study describes the development and background levels that negatively affect the application o f a reproducible method for assessing sensitivity o f the assay. the glycation level o f membrane proteins in mammalian tissue. To date, studies describing the Our strategies to overcome these problems include biochemical effects o f nonenzymic glycation in the the following: / ) The use o f CHAPS for membrane pathogenesis o f diabetes have mainly focused on protein extraction. Although Triton X 100 is often purified soluble proteins. Only a hancUul o f studies the preferred choice for extracting proteins fiom focus on membrane-associated proteins, mainly by membranes, CHAPS has a solubilizhig efficiency exposing isolated cell systems to high concentrations comparable to that o f Triton X 100, according to our o f glucose for prolonged periods o f time in vitro. SDS-PAGE and protein quantification data (not Using isolated erythrocytes, for example, Bilan and riiowtt); fir our experience; nmltiple washing and K lip (2) showed that ^ycation alters the maximal redissolving steps are not sufficient to completely transport rate (Jmax) o f the glucose transporter for remove Triton X 100 from the protein pellet, most cytochalasin B, a well-characterized competitive likely because of the tight intrinsic association of iithibitor of this carrier protein. Triton X 100 with membrane protein fiactions. 2) The use o f a desalting PD-10 column to eliminate The lack o f data on membrane-boimd proteins is ions and small molecules that we found caused high likely due to the absence o f a protein extraction background readings in the subsequent assay. S) The protocol in combination with a sensitive assay. The introduction o f a protein precipitation step using established method for measuring protein glycation TCA and a consequent redissolving step to obtain by Kennedy et al. (7) was originally based on several reproducible results. These treatments remove purified model proteins, i.e., bovine serum albumin, nonproteinaceous macromolecules that we found human immunoglobulin G and fetal calf fetuin. The intwfered with the assay. 4) The use of buffers initial protein isolation protocol contained one containing 1% SDS in 0.1 M sodium phosphate to AAPS Phamuel 1999; 1 (4) artlel* 20 (http:/Ai»ww.phanntei.era/)

— ~ redissolve~therprDtein peltetr" contain Tris or HEPES that may interfere with the I. Bakan E, Bakan N. Glyccsybiion o f nail m diabetics: possible marker o f assay because o f high absorbance o f these bng-tetmhyperglycemia. ClinChim Acta. I98S;I47:I-S. compounds at the assay detection wavelength. The 1 Bilan PJ, K lip A . Glycation o f the human erythrocyte glucose transporter in combination o f these steps, as illustrated in Figure 1, vitro and its fimctiotud consequences. Biochem J. I990;268:66l-d67. allowed us to reproducibly determine the glycation 3. Watkins N C , Thorpe SR, Baynes JW. Glycation o f amino gmiqrs in protein. level o f total apical membrane protein. Studies on the specificity o f modification o f RNase by glucose. J Biol Chem. I98S;260:10629-10636. Using the above method, we assessed the glycation 4. Cerami A, Stevens VJ, Momuer V M . Role o f nonenzymatic glycosylation in level o f total membrane protein from diabetic tissue the development o f the sequelae o f diabetes mellitus. MetabolisitL 1979:28:431 - to be 2.2-fold higher than that from control tissue. 437. This result is considerably lower than the ratio o f 3.2 5. Gallop PM , Ruckiger R, Hanneken A , Mininsohn M M , Gabbay KH. we detected for GHb levels in diabetic and control Chemical quantitation o f hemoglobin glycosylation: Ruorometric detection o f formaldehyde released upon periodate oxidation o f glycoglobin. Anal Biochem. animals, but further studies are necessary to frilly 1981;117:427-432. explain this phenomenon. Interestingly, Bilan and 6. Ahmed N , Furth AJ. A microassay for protein glycation based on the Klip (2) found glycation ratios (glycatedrcontrol) in periodate tnethod. Anal Biochem. 1991;192:109-111. erythrocyte glucose transporters ranging fipom 1.8- 7. Kennedy D M , Skillen AW, Self CH. Colorimetric assay o f glycoprotein ;2.8 following exposure o f isolated erythrocytes to glycation fire o f interference from glycosylation residues. Clin Chem. 80-;200 mM glucose for 3-;6 days. The Jmax value for 1993;39:2309-2311. this system was decreased to ~75% o f its control 8. Junod A, Lambert AE, Orci L, Pictet R, Gonet AE, Renold AE. Studies o f the value, which demonstrates the adverse effects o f diabetogenic action o f streptozotocin. Proc Soc Exp Biol Med. I9 6 7 ;I2 6 :2 0 I- 205. glycation on transport protein function. Future efforts are directed toward isolating single nutrient 9. Oremefc 0 , Seiffert UB, Schmid G. Determination o f glycated hemoglobin by afUnity chromatography. tZIin Chim Acta. 1987;168:81-86. transporters from the pool o f total membrane proteins to directly study the effects o f glycation on 10. Marshak, DR. Strategies for protein purification and characterization: A laboratory course manual. Cold Spring Harbor Laboratory Press, Plainview, N. transport function and expression. The currently Y. 1996. described glycation assay w ill serve as a benchmark I I . Smith PK, Krohn RL Heimanson GT, M allia AK, Gartner FH, Provenzano for assessing glycation levels in single proteins. M D, Fujimoto EK, Goeke NM , Olson BJ, Klenk DC. Measurement o f protein using bicinchoninic acid. Anal Biochem. 1985; 150:76-85. To our knowledge, the method described in this paper is the first to directly determine the degree o f 12. Laemmli UK. Cleavage o f structural proteins during the assembly o f the head o f bacteriophage T4. Nattne. 1970;227:680-685. glycation o f membrane proteins extracted from intestinal tissue using <100 pg protein. Because 13. Towbin H , Staehelin T , Gordon J. Electrophoretic transfer o f proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. membrane proteins play an important role in Proc Natl Acad Sci U S A 1979;76:4350-4354. transporting nutrients and drugs in the intestine and 14. Shneider B L, Dawson P A Christie D -M , Hardikar W , Wong M H , Suchy FJ. various other o r^ s in the body, the current protocol CIonitig.and molecular characterizatian o f the ontogeny o f a tat ile a l sodium- w ill be particularly useful to researchers dependent bile acid transporter. J Clin Invest. 1995;95:745-754. investigating the role of membrane protein glycation 15. Stravitz RT, Sanyal AJ, Padnak W M , Vlahcevic ZR, Beets JW, Dawson P A in the pathology o f aging and metabolic Æseases Induction o f sodium-dependent bile acid transpotter messenger R N A protein, and activity in rat ileum by cholic acid feeding. Gastroenterology. I9 9 7 ;I3 :I5 9 9 - such as diabetes. 1608.

ACKNOWLEDQMENTS 16. Christie D M , Dawson P A Thevananther S, Shneider B L Comparative analysis o f the ontogeny o f a sodium-dependent bile acid transpotter in rat The current work was funded through a new kidney and ileum. Am JPhysioL 1996;271 :G377-385. investigator award from the Pharmaceutical Research and Manufacturers of America Foundation (to P.W .S.). The authors wish to express their gratitude to Dr. John A. Bauer (Division o f Pharmacology, Ohio State University) for providing diabetic rat tissue used in this study.